 

A ..... 42..
v ’ a.

.‘I ‘I

\

.1:

r.

 

 

$44593. ,
deﬂuwtuu. at... ‘

.§

my,

“Mums? 41..
c raw

r,
‘ 5 is;
9;: .
3t :0

imam

inmwﬂi

, to...

. i... .
at!!! .l
nu”... as... w! 2.1!!
. faith.“ 3 .ii

til?»

i . .f ,

~ 3‘1 .41.

1.. ‘A ‘ 5.51.. .

Vial-yashhu“ \h!u...‘f“ﬂcma|.r I: 1'34““.

:3"! 2).. .hlfvﬁr)!\r$x)’ti’z
i I. {I 51,...(:

5;?! run.“

'1! . :1.
313...}: .55!
s

.

L11. .. huria.x.-.f
ﬁnkxltﬂouznci. .t‘

g‘l'lilialtt. lI-l L1,:("II

, r 7 I: .. i ltl. 1"! I! v.

t...“ 9.3.31.5.91 listilu (I.

c.
1.1.}! .
X’valnllllxll!
If . ‘Ilsxio...’
(I, .

11.1.“ «u‘

7‘52... M.

 

ThFSlS

‘ LIBRARY
Michigan State 3
University _

_l

 

This is to certify that the

dissertation entitled

An Investigation of the Effect of Harmonic Accompaniment

on the Developmental Tonal Aptitude, Tonal Achievement,

and Tonal Improvisations of Children in Kindergaten and
First Grade

presented by

Denise Marie Guilbault

has been accepted towards fulﬁllment
of the requirements for

Ph.D. degree in Music Education

 

 

 

CW; [ﬁrIM

0 Major professo'r‘ .3 A

DateJZ/lo !02

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

MAY 102004

 

JLCE I 8-1299"

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRClDatoDuopes-sz

 

AN INVESTIGATION OF THE EFFECT OF HARMONIC ACCOMPANIMENT
ON THE DEVELOPMENTAL TONAL APTITUDE, TONAL ACHIEVEMENT,
AND TONAL IMPROVISATIONS OF CHILDREN IN KINDERGARTEN
AND FIRST GRADE

By

Denise Marie Guilbault

A DISSERTATION
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
School of Music

2002

ABSTRACT
AN INVESTIGATION OF THE EFFECT OF HARMONIC ACCOMPANIMENT
ON THE DEVELOPMENTAL TONAL APTITUDE, TONAL ACHIEVEMENT,
AND TONAL IMPROVISATIONS OF CHILDREN IN KINDERGARTEN
AND FIRST GRADE
By

Denise Marie Guilbault

The purpose of this research was to examine the effect of harmonic
accompaniment on the developmental tonal aptitude, tonal achievement, and
tonal improvisations of young children. The specific problems of this study were
the following: 1) Does the addition of a root melody accompaniment to song
instruction affect the developmental tonal aptitudes of children in kindergarten
and ﬁrst grade? 2) Does the addition of a root melody accompaniment to song
instruction affect the tonal achievement of children in kindergarten and ﬁrst
grade? 3) Does the addition of a root melody accompaniment to song instruction
affect the tonal strength of the improvisations of children in kindergarten and ﬁrst
grade?

Sixty-eight kindergarten students and 68 first grade students (N = 136)
from eight intact classrooms participated in the study. All children took the Tonal
subtest of lntennediate Measures of Music Audiation as a pre- and post-test.
Instruction lasted for 25 weeks and consisted of rote song instruction with root
melody accompaniment (experimental group) and without root melody
accompaniment (control group). Other than the use of root melodies, all children

received the same instructional content. At the end of the 25-week instructional

period, children performed a major criterion song, a minor criterion song, and
improvised an ending to an unfamiliar song in major tonality. The song
performances were audio-taped and rated by three independent judges for
Tonality and Keyality. The improvisations were rated for Tonal Strength.
Results indicated that song instruction with a root melody accompaniment
had no signiﬁcant effect on the developmental tonal aptitude and tonal
achievement of Children in kindergarten and ﬁrst grade. As would be expected,
children in kindergarten received lower tonal aptitude mean scores and Tonality
mean scores than children in the ﬁrst grade. However, no signiﬁcant difference
was found between children in kindergarten and ﬁrst grade for Keyality mean
scores and Composite Tonal Achievement mean scores. Children who received
song instruction with root melody accompaniment achieved signiﬁcantly higher
Tonal Strength scores when improvising than children who did not have such
instruction. As would be expected, children in kindergarten scored signiﬁcantly

lower than the children in ﬁrst grade.

Copyright by
DENISE MARIE GUILBAULT
2002

This paper is dedicated to the memory of my father

Roland William Guilbault.

AKNOWLEDGEMENTS

If ever I have understood the phrase “no man is an island," it would be
now as I complete this document. The enormity of a project such as this
underscores the importance of each and every individual who has helped in one
way or another. Paul Torre, my friend and colleague from the Flint Institute of
Music in Flint, Michigan, planted the “Ph.D. idea” into my head. His insistence
and steadfast belief in me was the beginning of this extraordinary journey. I owe
you one Paul!

A most heartfelt thanks goes to Cheryl Barnes and Miriam Moore, my
friends and colleagues at Cranbrook Schools - Brookside in Bloomﬁeld Hills,
Michigan. They gave up personal curricular goals, and students, in order to help
with this study. My appreciation also goes to the school’s administrative staff,
teachers, students, and parents who were more than accommodating. Thanks
also go to Jennifer Bailey, Steve Lorenz, and Kristin Slocum-Kreiss for their
assistance, friendship, and generosity.

A special thanks goes to Joseph Lulloff for his contributions to this project;
to Dr. Judy Palac for offering her expertise and support to this study; and to Dr.
John Kratus for Sharing his unique and insightful views and for his mentorship
throughout my doctoral program.

I am forever indebted to my academic advisor, Dr. Cynthia Taggart. Her
scholarship, friendship, guidance, patience, and support were sources of

motivation and comfort for me. Cindy’s enthusiasm for this project was a driving

vi

force behind its fruition. I could not have done it without her. Cindy, I thank you
from the bottom of my heart.

Finally, it is with deep love, that I thank my mother “Tiny," my immediate
family, Annette, Bill, Bob, Leo, Paulette, Renee, my eight nieces and nephews,
and my life-long friend, Susan for not disowning me as a result of this endeavor.
They continued to support and encourage even after the all too frequent “sorry, I
can't - I have to work on the paper” episodes. Most of all, I give deep love and
thanks to my son Jacob Kane Langworthy. He is my inspiration and hope for the

future.

vii

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................... xi
CHAPTER ONE
RESEARCH BACKGROUND ....................................................................... 1
Introduction ................................................................................................ 1
The Importance of Developmental Music Aptitude .................................... 3
Children’s Music Improvisations ................................................................ 7
Need for Improvisation .................................................................... 7
Nature of Improvisation ................................................................... 9
Effect of Instruction on Improvisation ............................................ 19
Summary ...................................................................................... 21
Harmonic Context .................................................................................... 22
Summary ...................................................................................... 26
The Focus of This Research ................................................................... 27
Purpose and Problems of the Study ............................................. 27
Deﬁnitions ..................................................................................... 28
Limitations .................................................................................... 28
CHAPTER TWO
RELATED LITERATURE ............................................................................ 30
Effect of Instruction on Developmental Tonal Aptitude ............................ 30
Summary ...................................................................................... 47
Effect of Instruction on Tonal Achievement ............................................. 47
Characteristics of Singing Tasks .................................................. 49
Text Verses No Text ..................................................................... 53
Summary ...................................................................................... 59
Effect of Instruction on Improvisation ...................................................... 60
Summary ...................................................................................... 70
Effect of Harmonic Accompaniment on Tonal Achievement .................... 71
Summary ...................................................................................... 90
Chapter Summary ................................................................................... 92
CHAPTER THREE
DESIGN AND ANALYSIS ........................................................................... 95
Sample .................................................................................................... 95
Design ..................................................................................................... 97
Procedures .............................................................................................. 97

viii

Research Instruments ........................................................................... 104

Data Analysis ........................................................................................ 106
CHAPTER FOUR
RESULTS AND INTERPRETATIONS ................................................... 109
Reliabilities of Pre-Instruction and Post-Instruction Tonal subtest
of Intermediate Measures of Music Audiation ............................. 109
Reliabilities of Tonal Achievement Rating Scale ................................... 109
Reliabilities of the Improvisation Rating Scale ....................................... 112

The Effect of Root Melody Accompaniment to Songs on the
Developmental Tonal Aptitude of Children in Kindergarten

and First Grade ........................................................................... 113
Unpaired tTest of IMMA Pre-lnstruction Tonal Scores ............... 113
Means and Standard Deviations of Tonal subtest of IMMA

Post-Instruction Scores .................................................... 114
ANOVA of Post-Instruction IMMA Tonal Scores ......................... 115
Interpretation of IMMA Tonal Aptitude Scores ............................ 116

The Effect of Root Melody Accompaniment to Songs on the Tonal
Achievement of Children in Kindergarten and First Grade ......... 119
Means and Standard Deviations of Tonal Rating Scale —

Tonality Dimension .......................................................... 119
ANOVA of Tonality Scores ......................................................... 123
Interpretation of the Tonality Score ............................................. 125
Means and Standard Deviations of Tonal Rating Scale -

Keyality Dimension .......................................................... 126
ANOVA of Keyality Scores ......................................................... 129
Interpretation of the Keyality Scores ........................................... 131
Means and Standard Deviations for Major and Minor

Criterion Song Composite Scores .................................... 133
ANOVA of Major and Minor Criterion Song Composite

Scores .............................................................................. 135
Interpretation of Major and Minor Criterion Song Composite

Scores .............................................................................. 136
Means and Standard Deviations for Composite Tonal

Achievement Scores ........................................................ 137
ANOVA of Composite Tonal Achievement Score ....................... 138
Interpretation of Tonal Achievement Composite Scores ............. 139

The Effect of Root Melody Accompaniment to Songs on the
Tonal Improvisations of Children in Kindergarten

and First Grade ........................................................................... 141
Means and Standard Deviations for Improvisation Scores ......... 141
ANOVA of Tonal Improvisation Scores ....................................... 142
Interpretation of Improvisation Scores ........................................ 143

ix

CHAPTER FIVE

SUMMARY, IMPLICATIONS, AND RECOMMENDATIONS ............ 145
Summary of the Study ........................................................................... 145
Purpose and Problems ............................................................... 145
Design and Procedures .............................................................. 146
Results and Interpretations ......................................................... 147
Conclusions and Implications for Music Education ................................ 153
Recommendations for Future Research ................................................ 154
APPENDICES
Appendix A: UCRIHS Permission .......................................................... 160
Appendix B: Letter to Parents ................................................................ 161
Appendix C: Parental Consent Form ..................................................... 163
Appendix D: Song Repertoire ................................................................ 164
Appendix E: Sample of Senior Kindergarten Lesson Plans ................... 165
Appendix F: Sample of First-Grade Lesson Plans ................................. 168
Appendix G: Verbal Assent Script ......................................................... 171
Appendix H: Improvisation Antecedent Phrase ..................................... 172
Appendix I: Criterion Songs ................................................................... 173
Appendix J: Preparatory Sequences ..................................................... 174
Appendix K: Tonal subtest of Intermediate Measures of
Music Audiation Test Form ......................................................... 175
Appendix L: Tonal Achievement Rating Scale ....................................... 176
Appendix M: Improvisation Rating Scale ............................................... 177
REFERENCES ............................................................................................ 179

LIST OF TABLES

Table 1 — Teacher Assignments ........................................................................ 99
Table 2 - lnterjudge Reliabilities of Rating Scales ........................................... 111
Table 3 — lnterjudge Reliabilities of Composite Tonal Achievement Score ...... 112
Table 4 - lnterjudge Reliabilities of the Improvisation Rating Scale ................. 112
Table 5 — Groupings of Pre-Instruction Tonal Scores Between

Treatment Groups And Split by Grade .................................................. 113
Table 6 — Unpaired tTest of Pre-Instmction Tonal Scores Between

Treatment Groups Split by Grade .......................................................... 114
Table 7 - Means Table of Post-Instruction Developmental Tonal Aptitude

Scores ................................................................................................... 115

Table 8 — ANOVA Table of Posttest Developmental Tonal Aptitude

Scores ................................................................................................... 116
Table 9 .- Means Table of Tonality Scores ....................................................... 121
Table 10 -— Means Table of Composite Tonality Scores ................................... 122
Table 11 — ANOVA Table of Tonal Rating Scale: Tonality Dimension ............. 124
Table 12 — ANOVA Table of Tonality Composite Score ................................... 124
Table 13 — Means Table of Tonal Rating Scale: Keyality Dimension ............... 127
Table 14 — Means Table of Keyality Composite Scores ................................... 128
Table 15 - ANOVA Table of Tonal Rating Scale: Keyality Dimension ............. 130
Table 16 -- ANOVA Table of Keyality Composite Score ................................... 130

xi

Table 17 — Means Table of Major and Minor Criterion Song Composite
Scores ........................................................................................ 134

Table 18 — ANOVA Table of Major and Minor Criterion Song Composite ........ 136

Table 19 — Means Table of Tonal Achievement Composite Scores ................. 138
Table 20 — ANOVA Table of Tonal Achievement Composite Scores ............... 139
Table 21 — Means Table of Improvisation Scores ............................................ 142
Table 22 — ANOVA Table of Improvisation Scores .......................................... 143

xii

CHAPTER ONE

INTRODUCTION

Singing has long been regarded as central to the elementary school music
curriculum (Apfelstadt, 1984; Atterbury, 1984a, 1984b; Atterbury 8 Silcox, 1993;
Cassidy, 1993; Clegg, 1966; Cooper, 1995; Flowers 8 Dunne-Sousa, 1990;
Froehlich, 1977; Green, 1987; Hale, 1977; Kimble, 1984; Klinger, Campbell, 8
Goolsby, 1998; Levinowitz, Barnes, Guerrini, Clement, D’April, 8 Morey, 1998;
Mizener, 1993; Rutkowski, 1986; Smale, 1988). In fact, the National Standards
for Arts Education (1994) lists singing alone and with others as something all
students (K-12) should know and be able to do. Many believe that singing has
maintained this role in the curriculum because of its effect on (1) the
development of musical understanding (Bertaux, 1989; Regelski, 1981; Smale,
1988; Waddell, 1989); (2) music aptitude (Gordon, 1997a); and (3) children’s
ability to audiate (Gordon, 1997a). In addition, musical understanding can be
measured and assessed through children’s singing (Barnes, Flowers 8 Dunne-
Sousa, 1990; Froehlich, 1977; Hale, 1977; Levinowitz, Guerrini, Clement, 0’
April, 8 Morey, 1998; Rutkowski, 1996). When children sing, their performance
is a reﬂection of their understanding of music.

In light of this, the development of children’s vocal skills has been an
important focal point of music research (Clegg, 1966; Cooper, 1995; Flowers 8
Dunne-Sousa, 1990; Green, 1987, 1990, 1994; Grutzmacher, 1987; Froelich,

1976; Hale, 1977; Kelly, 1998; Kimble, 1984; Levinowitz, Barnes, Guerrini,

Clement, D'April, 8 Morey, 1998; Moore, 1994; Rutkowski, 1986; Rutkowski,
1996; Stauffer, 1986; Welch, Sergeant, 8 White, 1995/96). The implications of
these studies have helped general music teachers understand matters such as
the processes required for accurate singing, the relationship between age and
singing ability, the relationship between pitch discrimination ability and singing
accuracy, and the effect of vocal range, melodic direction, intervals, vocal model,
text, and unison/solo singing on children’s vocal accuracy.

The typical child has the potential to learn to sing in tune, in meter, and in
tempo (Atterbury, 1984b; Gordon, 1997c; Phillips, 1996); yet not all children
mature into adults who are able to do so. Researchers have uncovered
knowledge about some elements of Singing that helps teachers understand how
to facilitate children’s vocal accuracy. Some of these elements are pitch
discrimination, pitch production, pitch monitoring, and the motivation to sing
(Atterbury, 1984; Goetze, Cooper, 8 Brown, 1990). One factor that has not
received much attention from the research community is the effect of harmonic
accompaniment on children’s developmental tonal aptitude, singing achievement
and improvisation. Perhaps this is because most music programs for primary
children place more of an emphasis on pitch, melody, and rhythm and give
minimal attention to harmony (Bridges, 1966) and because such programs rarely
teach children to sing or improvise within the context of an underlying harmonic

structure (Gordon, 1998b).

The Importance of Developmental Mgsic Aptitrﬂe

Music aptitude is the measure of Children’s potential to learn music
(Gordon, 1980b, 1986, 1989a, 1997a, 1997c, 1999). Aptitude is sometimes
mistaken for achievement. Music aptitude represents a student’s potential to
learn, whereas music achievement represents what a student has Ieamed.
There are two classiﬁcations of music aptitude - developmental and stabilized.
Children’s music aptitude is developmental during the ﬁrst nine years of life.
During this time, the quality of the music environment can positively or negatively
inﬂuence children’s potential to achieve in music (Gordon, 1980b, 1986, 19893,
1997a, 1997c, 1999). The environment ceases to have an effect on aptitude at
about the time children are nine years old. At this point, aptitude becomes
stabilized (unaffected by the environment) and will remain so throughout the
child’s life.

Differing perspectives on the origins of music aptitude have fueled many
debates throughout the years. There are some who insist that music aptitude is
innate and unaffected by environmental factors, while others hold the view that
music aptitude is entirely environmental (Gordon, 1998a). Current ﬁndings in
psychomusicology suggest that music aptitude is a product of both innate
potential and musical environment (Gordon, 1998a). The proportion that nature
and environment contribute to music aptitude is still unknown. What is known is
that, regardless of aptitude level, children’s potential to achieve in music will not

be adequately developed unless they are exposed to a quality music

environment. In addition, the younger a child is, the greater the environmental
inﬂuence will be (Flohr, 1981; Gordon, 19803, 1980b; Taggart. 1997).

Numerous biologists, medical researchers, neurobiologists, neurologists,
pediatricians, psychologists, and scientists have suggested that there are critical
periods associated with the wiring of neurological connections and synapses that
take place prenatally and throughout early childhood (Begley, 1996; Kotulak,
1993; Nash, 1997). Before birth and at critical periods after birth, a baby’s brain
produces an abundance of cells that are used to make connections, or synapses,
between neurons (cells that carry electrical messages through the nervous
system and the brain). While the brain contains nearly all the nerve cells it will
ever need at birth, the pattern of wiring between them has yet to stabilize.
Sensory experiences create the circuits for abilities such as vision, sight, hearing,
smell, and language. A child’s brain development suffers if deprived of a
stimulating environment, because the brain eliminates those synapses that are
seldom or never used.

The discovery that the outside world is indeed the brain’s real food is truly

intriguing. The brain gobbles up its external environment in bits and

chunks through its sensory system: vision, hearing, smell, touch and taste.

Then the digested world is reassembled in the form of trillions of

connections between brain cells that are constantly growing or dying or

becoming stronger or weaker, depending on the richness of the banquet

(Kotulak, 1993, p. A8).

This applies to the development of the parts of the brain that processes music as
well.

Thus, if a very young child has no opportunity to develop a music-listening

vocabulary, the cells that would have been used to establish the hearing

sense will be directed to another sense, perhaps the visual, and so the
visual sense will be strengthened at the expense of the aural sense.

Regardless of one’s innate potential, no amount of compensatory

education at a later time will be completely able to offset the handicap

(Gordon, 1998a, p. 9).

The music environment of young children ought to be approached with great care
in light of its signiﬁcant effects on children’s developing music aptitudes and
future achievement.

The younger a child is when the child receives appropriate music
instruction, the more a child will be able to achieve in music (Gordon, 1997a,
1997c, 1998a). Gordon developed three valid measures of tonal and rhythm
developmental music aptitude that can be used to determine the strengths and
weaknesses of individual children. The scores from these tests can also be used
as a guideline when adapting instruction to the individual musical differences of
children. Audie (Gordon, 1989b) is a developmental music aptitude test for
three- and four-year-old children. A parent at home or a teacher in school
individually administers the test. The Primary Measures of Music Audiation
(Gordon, 1979) is a developmental music aptitude test for children in

kindergarten through third grade. It is called a test of audiation and not a test of

aptitude, because aptitude is still developmental for children in this age group.
The Intermediate Measures of Music Audiation (Gordon, 1982) was designed for
groups in which half or more of the students score above the 80th percentile on
the Tonal and Rhythm subtests of Primary Measures of Music Audiation.

Developmental music aptitude tests are extremely Important, because,
without them, teachers can only guess the potential of children based on their
achievement (Gordon, 1998a; Taggart, 1989). The problem with this is that
children sometimes underachieve. Music teachers cannot effectively evaluate a
child based solely on what is observed. Some children will have high
developmental music aptitudes and their musical behaviors will be
commensurate, and sometimes they will not (Gordon, 1998; Taggart, 1989). It is
through the use of music aptitude tests in conjunction with observation that music
teachers can objectively identify the potential of the children that they teach.
Only after objective measurement has been made of a child’s aptitude and
achievement in music can a teacher expect to make a dependable subjective
evaluation (Gordon, 1997a).

Because all children have the potential to achieve in music, andbecause
children’s potential for future achievement is inﬂuenced by the quality of the
music environment, it is important that children receive the best possible
instruction. It is imperative that teachers understand the signiﬁcance of
developmental aptitude to facilitate the development of their students potential to

the highest levels.

Children’s Music Improvisations

Need for Improvisation

The quality and extent of one’s early musical environment, which affects
one’s overall music aptitude, is perhaps the most powerful factor in determining
the extent to which one can become musically creative (Gordon, 1989a). When
a teacher provides students with the readiness and skills to create and improvise
their own music, music becomes the property of the students themselves, and
this should represent the ultimate goal of all music teachers (Azzara, 1992 8
1999; Gordon, 1997a). \lmprovisation means that an individual has internalized
(can audiate) a music vocabulary and is able to understand and express
intended musical ideas spontaneously.

Improvisation is to music what speaking is to language. Individuals

improvise daily with language when engaged in conversation. Like an

individual’s contribution to conversation, improvisation in music is

generated from an internal source. To understand music, one must think

musically. Improvisation skills allow students to express musical thought

and ideas from that internal source, with meaning (Azzara, 1992, pp. 6-7).

Although improvisation has been a vital part of music making throughout
history, it is inexplicably missing from most school music curricula today (Azzara,
1992). The importance of improvisation as a valuable music skill is noted in the
National Standards for Arts Education (1994), yet this is one of the standards that
is least-successfully implemented (Adderly, 1999; Kirkland, 1996). The third

content standard for children in kindergarten through fourth grade lists

improvising (1) musical “answers,” (2) simple rhythmic and melodic ostinato
accompaniments, (3) simple rhythmic variations and melodic embellishments,
and (4) short songs as a music skill children should know and be able to do.

Gordon (1997a) states that, as a spontaneous expression of musical
ideas, improvisation is a creative activity, but important differences exist between
"creativity" and "improvisation." Creativity involves fewer restrictions than
improvisation. Improvisation involves speciﬁc guidelines that provide structure
for the improviser — for example, the tonality. the harmonic progression, the
meter, and the form of the music. In addition, improvisation requires more
complex audiation than creativity (Gordon, 2001 ).

Tonal improvisation may take several forms. Tonal improvisation may
include a performance of a variation of a melody. This requires that the student
can memorize and imitate. Awareness of tonic, resting tone, or harmony is
helpful but not essential. Tonal improvisation may be the performance of a
melody over a given harmonic progression, or it may be the improvisation of
harmonic progressions to a melody. Both of these approaches require audiation
skills. Tonal improvisation is most meaningfully accomplished when the
improvisor is audiating a harmonic context (Gordon, 1997b 8 1998b).

Experiences with improvisation have been shown to have a positive effect
on student's other musical skills. The purpose of Azzara’s study (1992) was to
enhance the music achievement of elementary instrumental music students
through a researcher-designed improvisation curriculum. The speciﬁc problems

of his investigation were to examine (1 ) the effect of improvisation on music

achievement and (2) the effect of levels of aptitude on music achievement.
Signiﬁcant differences existed between students who received instruction that
involved the use of the improvisation curriculum and those who did not. The
treatment group received higher ratings on their performances of three criterion
etudes than students in the control group. As expected, signiﬁcant differences
were also found between aptitude levels. The group of students who were
classiﬁed as having high music aptitude demonstrated higher achievement than

those students with low music aptitude.

Nature of Improvisation

Research has shown that improvisational and compositional processes
and products can be used to evaluate musical understanding (Briggs, 1987;
Brophy, 1999; Cohen, 1981; DeLorenzo, 1989; Doig, 1941, 1942a,b; Flohr, 1980,
1981,1985; Fruendlich, 1978; Joseph, 1983; Kratus, 1985, 1986, 1989, 1991a,
1991b, 1991 c, 1994, 1995, 1996; Laczo’, 1981; Montano, 1983; Moorhead 8
Pond, 1978; Martin, 1993; Munson; 1986; Pond, 1981; Prevel, 1979; Reinhardt,
1990; Webster, 1989). Children can and do improvise and compose in
meaningful ways. Improvisation is the focus of the present study; therefore,
composition will not be discussed.

A few researchers have examined children's tonal improvisations. During
the 1940’s at the Pillsbury Foundation School in Santa Barbara, California,
Moorhead 8 Pond (1978) published one of the ﬁrst naturalistic studies on

children's improvisations. They observed and recorded the spontaneous singing,

chanting, and instrument playing of children, age two through six. Moorhead and
Pond found that, with guidance and experience, young children could (1)
improvise tonal and rhythm patterns, (2) improvise with a steady beat, and (3)
improvise asymmetrical patterns. Early attempts at instrumental improvisations
primarily involved tone and timbre explorations.

Prevel (1979) examined over 2,000 children’s spontaneous improvisations
in a free and unstructured environment. His observations suggest that children’s
motor energy is reﬂected in their ﬁrst attempts at improvisation. What is most
important to children is that they identify with the overall effect of their own music
making. When children gain the ability to restrain their own movements, they
experiment with different colors of sound and different volumes of sound. They
also create accents, conclusions, and even introductions. He concluded that the
development of children’s improvisations owes more to their kinesthetic
development than their auditory perception.

Flohr’s research (1980) suggests that young children can meaningfully
improvise music that incorporates simple musical structures. Flohr observed the
improvisational behaviors of four-, six-, and eight-year-old children over a period
of ten 15—minute individual sessions. The children improvised on an Orff
xylophone using a two octave pentatonic scale. Each child individually
participated in three improvisatory phases. The ﬁrst phase was free exploration.
The children were given ﬁve minutes to freely explore the sound possibilities of
the instrument. The second phase, guided exploration, consisted of a series of

eight researcher-directed tasks. Phase three was exploratory improvisation.

lO

Flohr performed “Hush Little Baby,” and then asked the children to make up a
song. Flohr played a bordun accompaniment on the bass xylophone while each
child improvised.

Flohr concluded that (1) young children could freely improvise music that
employs musical patterns and complex rhythms; (2) the characteristics of
children’s improvisations change in relation to chronological age; (3) older
children maintain focus on the sound possibilities of a single instrument longer
than four-year-old children; (4) a sense of tonality and consistency in
improvisations become more evident as children mature; (5) young children can
improvise patterns that are related to musical stimuli; and (6) young children are
able to create musical images in response to verbal stimuli.

In a 1985 study, Flohr characterized and described the behavior of four
different age groups (ages two through ﬁve) engaged in improvisatory tasks. Ten
children from each of the four age groups served as subjects (N = 40). Flohr met
with the children each year for four years. Four of the children participated in the
study for all four years. Eight children participated for three of the years, seven
children for two years, and 10 children participated for one year. As in Flohr’s
1980 study, each child individually participated in a three-phase interview.

The results of Flohr’s study led him to conclude that there are differences
in children’s improvisations that are related to chronological age. He proposed
three stages of improvisational development. Stage one, motor stage, is

characterized by slow and accented durations. He found that the two-year-olds

11

were more interested in the sounds the instrument made and were motivated
primarily by the physical movement involved in the production of sound.

In stage two, experimentation, children experiment with their ability to
produce a variety of sounds. Children tend to focus on motives exclusively,
explore many techniques for sound production, imitate the bordun rhythm, and
use quarter-note triplets, asymmetrical phrases, stepwise motion, and repeated
pitches. Sequences, inversions, variations, and repetitions can occasionally be
found in stage two improvisations.

Improvisations in stage three, formal properties, reﬂect the child's
understanding of musical form; improvisations end with short codas and
correspond to the length of the bordun accompaniment. Additionally, children
showed a preference for the pentatonic scale centered on the f tonality or the
related minor tonality of d.

Reinhardt (1990) examined preschool children’s use of rhythm in
improvisation in order to describe the rhythmic elements found in the children’s
improvisations and to determine if there were significant differences in the use of
rhythmic elements among the age groups. Analyses of the rhythmic
characteristics show that children were able to establish and maintain a steady
pulse and to establish and maintain the meter. Reinhardt proposes that the
ability to improvise with a steady beat and meter is stabilized by the time children
are three years old. Five-year-old children had the highest number of different
durations, and the three-year-old children had the least amount of different

durations. The differences were statistically signiﬁcant in both cases. Reinhardt

12

suggests that the use of different durations is in development between the ages
of three and ﬁve. The number of improvisations that did not contain a rhythm
pattern was highest for the three-year-old children and lowest for the ﬁve-year-
old children. Extended rhythm patterns were observed less often for the three-
year-old children and most often for children five years of age. Again, Reinhardt
suggests that the use of rhythm patterns in improvisation develops between the
ages of three and ﬁve years.

Freundlich (1979) conducted two case studies of ﬁfth grade boys in order
to examine the development of musical thinking. The two boys improvised using
a diatonic xylophone in the context of a 12-bar blues progression in the keyality
of A. Freundlich held “jam sessions” with the individual students once a week for
11 weeks. The children improvised on the xylophone while Freundlich performed
a rock and roll accompaniment on the guitar.

Freundlich concludes that children can produce musical ideas without
relying upon notation. On the other hand, the “rhythmic buoyancy” of the blues
progression may have “inhibited the child’s ability to deal successfully with
melodic-harmonic issues” (p. 59).

This study is troublesome on several accounts. First, Freundlich classified
his study as longitudinal when it consisted only of 11 weekly ﬁve-minute
sessions. Second, the use of an A major blues accompaniment with a diatonic
xylophone pitched in C seems ill-advised. Freundlich Should have replaced the
tone bars so that the xylophone could play in the same key as the guitar

accompaniment. Perhaps this, rather than the “rhythmic buoyancy” of the blues

13

progression, inhibited the boys’ abilities to improvise successfully in a harmonic
context. Third, little information was given about the subjects in this study.
Background information and interviews or discussions with the students might
have provided insight into their thought processes when improvising.

Brophy (1999) studied the instrumental (alto xylophone) improvisations of
280 six- through twelve-year-old Children. The children were asked to improvise
on the B, C, and D sections of a rondo. The analysis of individual melodic
features suggest that the use of speciﬁc ranges and intervals, the use of speciﬁc
type of melodic motion, and the use of the tonic as a closing tone were not
related to age (76% of the children did not end on the tonic). The use of
repeated and developed melodic motives increased signiﬁcantly with age. Older
children tended to experiment with their improvised melodic motives while
younger children repeated a melodic motive (or used no motive at all). Overall,
melodic features show signiﬁcant differences among age levels.

The analyses of individual rhythmic features suggest that the use of
rhythmic divisions were related to age. The presence of beat divisions that form
rhythmic patterns signiﬁcantly increased with age. Younger children improvised
melodies that were less consistent in terms of a steady pulse than older children.
Nine-year-olds were the most consistent with tempo. Signiﬁcant differences
were also found between the children’s use of repeated and developed rhythmic
motives. Older children repeated and developed rhythmic motives more
frequently than younger children. Signiﬁcant change did occur with age, which

suggests dynamic stages of rhythm development. The development of the

14

combined rhythmic features occurs from ages six through nine, as seen by the
steady progression toward the use of conventional rhythmic features in
improvisations. Between the ages of nine and ten, there appears to be a
developmental plateau, in which the growth temporarily ceases but reappears at
age 11 and continues through age 12.

The analyses of the structure features for the presence of phrase suggest
signiﬁcant age-related differences. Older children improvised melodies that were
structured in phrases more often than younger children. Also, older children
improvised melodies that were structured in antecedent/consequence phrases.
Overall, results show signiﬁcant differences between and among age levels,
although these changes did not exhibit a consistent increase as the differences in
age level increased. The differences in the use of structural features between
ages six and nine, between eight and nine, and also between ages eight and 12
were signiﬁcant. No signiﬁcant difference was found between ages eight and
ten/11. The presence of structural features increases signiﬁcantly between ages
six through nine, with the most rapid growth occurring between ages eight and
nine. There is a decline in the presence of structural features at age ten, and
then a resumption of growth at ages 11 through age 12.

The three Total Improvisation Scores provided a measure of the musical
characteristics of each of the three improvisations as an eight-measure musical
entity. There were signiﬁcant differences between and among age levels for

each of the three Improvisations. Significant differences were found between

15

ages six and seven, ages nine through 12, and between ages eight and nine for
the third improvisation.

The Overall Composite Score represented the three improvisations as a
24-measure musical product. The results suggest that the combined
characteristics of the improvisations change signiﬁcantly with age. The scores
signiﬁcantly increase between ages eight and nine, and this increase serves to
partition the six- through eight-year-olds and the nine- through 12-year-olds into
two signiﬁcantly different groups. Brophy suggests that a dynamic stage may
exist in the development of the overall characteristics. The first dynamic stage
can be identiﬁed as ages six through nine. During the ﬁrst stage, the six-,
seven-, and eight-year-olds progress steadily toward a musical conformity in their
improvisations, with a period of rapid growth between ages eight and nine. A
developmental plateau occurs for ages nine through 11, with growth resuming at
age 12.

Munsen (1987) investigated children’s (grades one, three, and ﬁve) ability
to improvise rhythmically and melodically (vocally and instrumentally) within a
question and answer format, in the context of an Orff-Schulwerk-based music
program. The children’s improvisations during the regular music class were
tape-recorded and rated with regard to rhythmic/tonal security, phrase length,
rhythmic/melodic ﬁnal point, musical congruence, rhythmic/melodic organization,
originality, and overall musical impression. Munsen concluded that there was a
noticeable difference between the improvisations of children in grades one and

three. Children in grade three received higher ratings for their improvisations

l6

than children in the ﬁrst grade. This was due in part to maturation and also in
part to the fact that children in ﬁrst grade had less opportunity to improvise than
had children in the third grade. Little difference was found between the
improvisations of children in grades three and ﬁve. Again, because of the
instructional format, children in third grade received more opportunities to
improvise than did children in the fifth grade.

Martin (1993) replicated Munsen's (1987) study. Martin found that
children in ﬁrst grade were rated the highest of the three grade levels on their
melodic improvisations. Martin suggests that one possible explanation for the
high ﬁrst-grade melodic improvisation scores is that the improvisations of these
children were quite structured (consisting of sol-mi-Ia melodies), while the ﬁfth-
grade students were encouraged to display originality. Children in the third and
ﬁfth grade were aware of the eight-beat phrase length and beginning and ending
notes for a particular tonality. Children in the ﬁfth grade received the highest
scores for rhythmic improvisations. In addition, fifth—grade students received the
highest scores for overall musical impression.

Laczo’ (1981) investigated the psychological aspects of the improvisation
abilities of lower and upper grade levels of elementary school children. The
problems of his study were to 1) identify types of melodies improvised by
different ages and types of education, 2) identify factors that inﬂuence the
achievement of improvisation of different ages and types of education, 3) see if

differences existed in improvisations with text and without text, and 4) identify

17

connections between the musical activity and passive recognition as measured
by the Seashore test.

Children from three schools in Budapest, Hungary participated in the
study. One of the schools had classes specializing in music using the Kodaly
philosophy. Another school specialized in foreign language. The third school
held normal classes, and all children attended two music Classes each week.
Data collection consisted of two tasks: 1) to vocally improvise a melody using the
text of a short poem and 2) to vocally complete a melody (without text).

Laczo’ found that the types of melodies of the most advanced
improvisations with text had a sense of tonality, relatively correct intonation, well-
structured musical material, and accurate orientation in the rhythmical-time
relations. The types of melodies of the most advanced improvisations without
text had a sense of tonality, tone-system and formation, and cohesion between
the question and answer. The majority of improvisations were in pentatonic.
Students in the lower classes achieved better rhythmic accuracy in without-text
improvisations. With regard to rhythm, students performed with greater accuracy
in their improvisations with text regardless of the school attended. Students in
the lower and higher classes achieved better intonation accuracy in without-text
improvisations. The with-text improvisations of lower school children in the
language and normal classes were better than the text improvisations of lower
school children in the specialized music classes. Overall achievement in both
text and no text improvisations was higher for students in the specialized music

classes, regardless of age, than in the other two types of classes. The scores

18

students received on the Seashore measure were not predictive of their
achievement in improvisation. Students who received low aptitude scores
improvised at higher levels than would be expected. Laczo’ concludes, “the
musical education and the accumulated musical experiences are decisive in the
achievement of improvisation” (p.41 ). In addition, “intensive musical education

has priority over chronological maturity. Achievement depends on skill” (p. 42).

Effect of Instruction on Improvisation

It is important that children have the readiness and skills to improvise their
own music. Music educators need to know the most appropriate and productive
means to help children develop the ability to improvise. The following section
presents studies that examine the effect of instruction on children's
improvisations. These will be discussed in more depth in chapter two.

Wig (1981) examined the Viability of music composition strategies in
teaching sixth- and eighth-grade band students’ (N = 28) to improvise. Wig found
that signiﬁcant gains were made in improvisation ratings after the composition
instruction.

Joseph (1983) sought to determine if a kindergarten music program based
on Dalcroze Eurhythmics would develop children’s instrumental improvisational
skills. She proposed that children receiving improvisation instruction would
outperform children who did not receive improvisation instruction. Joseph found

that children receiving Eurhythmics training with improvisation instruction used

19

more patterns when improvising on a set of chromatic bells than those children
not receiving improvisation instruction.

Kalmer and Balasko (1987) examined the improvised songs of Hungarian
preschool children (N = 30) to determine if the music they were learning in their
classes created a musical vocabulary (referred to as “mother tongue”) that was
reﬂected in their vocal improvisations. Kalmer and Balasko suggest that the
“musical mother tongue" (p.83) does exist. Typical features of the Hungarian folk
songs (volume, tonality, intervals, structure, phrases, rhythm patterns) were
identiﬁable in the children’s improvisations.

Amchin (1995) investigated the effects of teacher/student verbal
interactions on the improvisations of fourth- and ﬁfth-grade students (N = 129).
He sought to “measure whether exposure to improvisation activities found in the
Orff-Schulwerk along with teacher-student interactions designed to help children
evaluate their creative responses inﬂuences children to produce more musical
improvisations” (p. 8). He found no signiﬁcant difference in the improvisations of
children who had had teacher-led discussions about their improvisations and
those children who did not.

Partchey (1974) examined the effects of three learning strategies on the
ability of sixth-grade students to improvise melodies. Students were randomly
assigned to one of three groups: 1) delayed feedback - the students immediately
heard a replay of their improvisation before improvising the second time; 2)
precomposed models - students heard a correct model in expectation that it may

positively inﬂuence subsequent improvisations, and; 3) repetition or control group

20

- received the same number of opportunities to improvise but did not experience
one of the two experimental strategies. This group served as the control for the
effect of practice on the pretest/posttest scores.

Partchey found no significant pretesUposttest mean score gains between
the groups. Statistically signiﬁcant pretest/posttest differences occurred between
all groups in the creativity and performance-creativity combined scores. Overall,
Partchey concluded that all three strategies do induce some measure of learning
as manifested in the ability to improvise melodies. All three strategies improved

improvisational ability.

Summary
Brophy (1999), Flohr (1980, 1985), Freundlich (1979), Martin (1993),

Munsen (1987), Moorhead 8 Pond (1978), Prevel (1979), and Reinhardt (1999)
show that children can improvise on an instrument using repeated and developed
melodic and rhythmic motives. They can also establish and maintain a steady
beat, maintain meter, maintain tonality, organize structural features (introductions
and codas), use different colors of sound, use different volumes of sound, and
use accents when improvising. Laczo’ (1981 ) found that children could
improvise vocally with a sense of tonality, relatively correct intonation, well-
structured musical material, and accurate orientation of rhythmical-time relations.
Amchin (1995) found that instruction that included verbal interaction did not affect
children's improvisations. Partchey (1974) found that improvisational skills

improved regardless of the learning strategy used.

21

Little research exists on young children’s vocally performed
improvisations. Furthermore, no research exists on the effects of vocal
instruction that includes a harmonic context on improvisational abilities.
Research is needed on the effects of a harmonic context on the vocal
improvisations of young children. When it comes to children having the
necessary readiness to improvise, it may be familiarity and comfort with chord

changes that is the compelling factor (Gordon, 2001).

Harmonic Context

How might teachers guide students toward familiarity and comfort with
Chord Changes? Students can broaden their music vocabulary by listening to
and learning music by ear. The process of Ieaming by ear and developing
musicianship begins at birth when children hear the music of their culture
(Gordon, 19970). In addition to Ieaming the melody, students should learn the
bass line and the harmonic progression for numerous tunes. Listening to bass
lines help students develop a melodic sense of harmonic progression, that is, a
sense of how melodies combine to create harmony. Azzara (1992, 1999)
suggests that students should learn to sing and play the bass lines of tunes by
ear, as he believes that knowing many melodies and bass lines is an invaluable
resource for improvising. As a result of learning songs and bass lines, students
will develop an understanding of how tones function within the harmony.
Students may also sing and perform with more accuracy. Bass lines provide

tonal context and meaning for musical performance.

22

There are differing Views as to the best techniques for helping students
perceive harmonic functions. Ritchie (1962) was interested in determining, with
sophomore college students (N = 58), whether the aural perception of melodic
fragments was enhanced or inhibited by the use of harmonic accompaniments.
He designed a 95-item test to compare the aural perception of various melodic
fragments presented as single lines with the aural perception of the same
melodic fragments appearing as the upper voices of four-part diatonic harmonies.

The melodic fragments were classiﬁed as belonging to one of two types —
diatonic (stepwise movement) or triadic (intervals of a minor third and larger).
Each melodic fragment was harmonized in a variety of ways. The
harmonizations were classiﬁed as follows: “Class I — triadic fragments using only
primary triads for the harmonization; (2) Class II — triadic fragments using any
triad or triads other than primary triads for harmonization; (3) Class III - diatonic
fragments using only primary triads for the harmonization; Class IV - diatonic
fragments using any triad or triads other than primary triads for harmonization”
(p. 14).

Ritchie found that (1) harmonization tended to decrease the accuracy of
aural perception, (2) harmonization using secondary triads decreased accuracy
of aural perception to a greater degree than harmonization consisting entirely of
primary triads, and (3) that the number of chord changes did not seem to be a
factor in the perceptibility of melodic fragments.

In order to identify the most effective means of teaching seventh- and

eighth-grade general music students (N = 106) to identify primary harmonic

23

functions, Alvarez (1980a) compared two harmonic classiﬁcation systems -
scalar and root. The scalar harmonic classiﬁcation system was used to identify
primary harmonic functions by classifying the progressions into one of three
classes using only the ﬁrst, seventh, and second scale degrees — tonic: I, IV, vi,
dominant: V, vii, iii, and pre-dominant: ii. For example, the chord progression of
l-iii-vi-ii-V-l in the tonality of C major using the scalar system would have a bass
line consisting of the pitches C-B-C—D-B-C. The root harmonic classiﬁcation
system was used to identify primary harmonic functions by classifying the
progressions into one of seven classes using all diatonic scale degrees. For
example, the chord progression of I-iii-vi-ii-V-l in the tonality of C major using the
root system would have a bass line consisting of the pitches C-E-A—D-G-C.

There was no signiﬁcant difference in the ability to identify primary
harmonic functions using root position chords between students who learned the
scalar system and students who learned the root system. However, students
who received instruction using the scalar technique were more adept at
identifying primary harmonic functions than students who were taught the root
technique. Alvarez concluded that the scalar technique is more effective than the
root technique in teaching seventh- and eighth-grade students to identify primary
harmonic functions.

In another study, Alvarez (1980b) investigated the most effective means of
teaching college students (N = 48) enrolled in one of four ear-training courses to
identify harmonic functions. Again, two harmonic classiﬁcation systems were

used to teach harmonic identification - scalar and root. The intact classes were

24

randomly assigned to either a scalar or a root classification system. Each class
received ten 50-minute sessions of instruction.

Results of the Alvarez (1980b) study show that the scalar classiﬁcation
system yielded signiﬁcantly higher mean scores on both subtests than did the
root Classiﬁcation system and, therefore, was the more effective means by which
to teach the identiﬁcation of harmonic functions. Alvarez states that the results
may be due to the following: (1) the scalar system consisted of three relevant
attributes (scale degrees one, seven, and two) whereas the root classification
system consisted of seven relevant attributes (all diatonic scale degrees); (2) all
relevant attributes of the scalar system were within the interval of a second of the
tonic pitch, whereas the relevant attributes of the root system encompassed a
larger interval above and below the tonic pitch; (3) the intervals of the relevant
attributes in the scalar system did not have to be inverted, whereas ascending
and descending intervals of the relevant attributes in the root system frequently
had to be inverted; and (4) the scalar system group was able to use the same
procedure for both root position and inverted harmonic functions, whereas the
root groups had to determine whether the perceived bass tones were actual roots
or inverted chord members.

Although older children and college students may be more successful at
identifying harmonic functions using the scalar technique, young Children may
need a different type of instruction. Children’s ability to audiate grows from their
ability to differentiate between sounds that are the same and not the same

(Gordon, 1997c, 1998a). Children who have listened to a wide variety of sounds

25

may develop a more sensitive ear than children who have not had the same
listening experience. This idea may apply to harmonic perception as well. The
scalar approach uses only three classiﬁcations, and therefore has limited
possibilities for making harmonic function comparisons. The root technique uses
seven classiﬁcations, and, as a result, may provide young children with more
opportunities to differentiate among and ultimately understand harmonic

functions.

Summary

The results of the above studies are insufficient to determine the best
venue for helping students develop an understanding of harmonic progression.
Perhaps the use of primary triads best serves this goal (Ritchie, 1962). It is also
possible that the scalar technique is a superior method (Alvarez, 1980a, 1980b).
Although research suggests that the root technique is inferior to the
aforementioned techniques (Alvarez, 1980a, 1980b), more research is needed
before a solid conclusion can be made. None of the studies mentioned above
involved young children. In relation to early childhood, the root technique may be
preferable. Since the root technique is the method suggested by Azzara, (1992)
and there is insufficient data to either conﬁrm or refute its effectiveness, further

research is needed so that ﬁndings are conclusive.

26

The Focus of This Research

The effect of instruction on developmental tonal aptitude, tonal
achievement, and tonal improvisation lies at the core of the present study. Music
instruction during the ﬁrst nine years of life can determine the extent to which
children give meaning to and derive pleasure from music throughout the
remainder of their lives (Gordon, 1997c, 1998a). Singing is an important vehicle
for supporting children’s tonal aptitudes and measuring children's understanding
of music syntax. Researchers have devoted much energy to unearthing the
instructional factors that are most beneﬁcial in supporting the development of the
young child’s singing. One instructional factor that has not received much
consideration is the effect of harmony on children’s singing as well as on their
developmental tonal aptitude. Furthermore, there are no studies that investigate

the effect of harmonic accompaniment on young children’s vocal improvisations.

Mose and Problems of the Study

In order to improve music instruction and to deﬁne factors necessary for
creating the optimal environment for music development, the purpose of this
research is to examine the effect of root melody accompaniment on the
developmental tonal aptitude, tonal achievement, and tonal improvisations of
young children. The speciﬁc problems of this study are the following:

1) Does the addition of a root melody accompaniment to song instruction
affect the developmental tonal aptitudes of children in kindergarten and

first grade?

27

2) Does the addition of a root melody accompaniment to song instruction
affect the tonal achievement of Children in kindergarten and ﬁrst grade?

3) Does the addition of a root melody accompaniment to song instruction
affect the tonal strength of the improvisations of children in kindergarten

and ﬁrst grade?

Deﬁnitions

Developmental Music Aptitude - The potential to achieve in music. It is affected
by the environment (until children are approximately nine years of age)
and can be measured by Intermediate Measures of Music Audiation.

Root Melody Accompaniment - The melodic line created by the fundamental
pitch of the harmonic functions found in the songs.

Tonal Strength — The degree to which children maintain a tonality and keyality
when improvising.

Tonal Achievement — Children’s ability to sing tonic, dominant, subdominant, and
cadential patterns accurately within the context of a major and minor song,

as well as to remain in the given keyality of the songs.

Limitations
1. Although there are many means through which to measure children’s tonal
achievement, the present study will be limited to measuring children’s

ability to sing tonic, dominant, subdominant and cadential patterns

28

accurately within the context of a major song and a minor song as well as
to remain in the given keyalities of the songs.

2. Although there are many means through which to measure children’s tonal
improvisations, the present study will be limited to children's abilities to
maintain a tonality and keyality when improvising.

3. Generalization of the results of this study is limited by the specific
characteristics of the children and the school in which the study was

conducted.

29

CHAPTER TWO

RELATED LITERATURE

The focus of this study is on the effects of root melody accompaniment
with song instruction on the developmental tonal aptitudes, tonal achievement,
and tonal improvisations of young children. Therefore, the literature that is
directly related to this study falls into four categories. They are the following: (1)
studies that consider the effect of instruction on developmental music aptitude,
(2) studies that consider the instructional factors that affect children’s singing
abilities, (3) studies that consider the effect of instruction on children's
improvisations, and (4) studies that consider the effect of harmonic
accompaniment on children’s music development. This review comprises a

discussion of each of those categories of study.

The Effect of Instruction on Developmental Tonal Aptitude
Several researchers have examined the effect of music instruction on
developmental tonal aptitude (Jordan-DeCarbo, 1982; Levinowitz, 1987; Lange,
2000; Kane, 1994; DiBlassio, 1985; Holahan, 1983; Jessup, 1984; Rutkowski,
1986; Rutkowski, 1996; Flohr, 1981; Taggart, 1997; Gordon, 1980a; Gordon,
1980b). Gordon’s Music Learning Theory states that tonal and rhythm aptitude is
developmental and ﬂuctuates until approximately age nine, depending on the

quality of the music environment and instruction (Gordon, 1997a, 1997c). The

30

following section includes studies that corroborate the theory that instruction
strengthens and supports developmental tonal aptitude. Still, there are
contradictory views as to the inﬂuence of instruction on developmental music
aptitude. These are presented first in this discussion.

Jordan-DeCarbo (1982) investigated the effect of training using
same/different discrimination techniques on kindergarten children’s aural
perception and vocal performance of tonal patterns and a song. Four questions
were addressed in her study: (1) Does involving Children in a decision—making
process (discerning same or different) affect aural discrimination and singing of
selected tonal patterns? (2) Does gender of kindergarten children interact with
aural ability as measured by aural discrimination (Primary Measures of Music
Audiation) and singing? (3) Does readiness training that includes the singing of
songs and resting-tone recognition inﬂuence aural ability as measured by aural
discrimination (PMMA) and singing skills? (4) Does reinforcing tonal patterns
with songs that provide a tonal framework inﬂuence aural discrimination (PMMA)
and singing? The following discussion will only include information relevant to
the ﬁrst, third and fourth problems, as they are the most closely related to the
present study. In this study, Jordan-DeCarbo uses the term aural discrimination
synonymously with music aptitude.

Five intact classes of kindergarten children (N = 89) from three different
elementary schools were randomly assigned to one of two groups. Children in
experimental group one received tonal pattern instruction using the

same/different technique of asking children to aurally discriminate between two

31

patterns. The children in the second experimental group received the same tonal
pattern instruction but responded through imitation and repetition rather than
aural discrimination. Children received three 20-minute periods of instruction
each week for 11 weeks. Jordan-DeCarbo administered the tonal subtest of
PMMA, as well as two researcher designed singing achievement tests.

Results of this study show there was no signiﬁcant difference in the tonal
subtest scores of PMMA or in the singing achievement test scores between the
two experimental groups. However, there were signiﬁcant gains from pretest to
posttest on both the tonal subtest scores of PMMA and the singing scores,
regardless of treatment. Jordan-DeCarbo was unsure as to whether the gain
was due to the effect of training or maturation.

Levinowitz (1987) conducted a study to determine the effects of song
instruction with words and without words on the levels of developmental tonal
aptitude and singing achievement of Children in kindergarten and ﬁrst-grade.

The Tonal and Rhythm subtests of Primary Measures of Music Audiation were
administered to children at the beginning and end of the academic school year.
The children were randomly assigned to one of three treatment groups: (1) songs
primarily with words, (2) songs primarily without words, and (3) songs always
with words.

Levinowitz found that mean gain scores for the Tonal and Rhythm
subtests of PMMA of children with low aptitudes were signiﬁcantly higher than
those of children with high aptitudes, regardless of treatment group. It is possible

that singing, regardless of whether one uses text, raises the aptitude of low

32

aptitude children. However, this could also be a result of regression toward the
mean. There was not a signiﬁcant difference between the pretest and posttest
scores of children with high aptitudes for any of the three groups. No signiﬁcant
differences were found in the tonal achievement scores between children in the
two experimental groups or between children with low and high developmental
tonal aptitude.

A portion of Lange's (2000) study was designed to consider the use of text
in song instruction and its possible effect on the developmental tonal aptitudes of
children in kindergarten. She also examined whether the use of text in song
instruction would affect Children’s ability to identify and perform the resting tone.

Children from four intact kindergarten classes from two different
elementary schools served as subjects for the study. Two classes (one from
each school) were randomly selected to receive song instruction primarily with
text for a period of 24 weeks. During the same 24 weeks, the remaining two
classes received song instruction primarily without text.

Using the children’s gain scores from PMMA, Lange found no signiﬁcant
difference between the tonal aptitude scores of the two treatment groups.
However, Lange did ﬁnd statistically signiﬁcant effects as a result of tonal
aptitude level. As expected because of the phenomenon of regression toward
the mean, children with low tonal aptitude had higher gain scores than did
children with high tonal aptitude. However, although the difference was not
statistically significant, the low tonal aptitude children who received song

instruction without text had higher gain scores than low tonal aptitude children

33

who received song instruction with text. There was no signiﬁcant difference in
the song or resting tone performances of children with low and high aptitudes
regardless of treatment group.

A portion of Kane’s study (1994) examined the effects of tonal pattern
instruction and song instruction on the developmental tonal aptitudes of
kindergarten children. She was speciﬁcally interested in the difference between
instruction that included the singing of tonal patterns and songs without words in
a variety of tonalities and meters (experimental treatment) and instruction that
included the traditional teaching of rote songs with words in major and harmonic
minor and duple and triple meters (control).

Eighty-seven children from four intact classes participated in the study.
Two of the four classes were randomly assigned to serve as the experimental
group. They received 30 weeks of instruction that included tonic and dominant
tonal patterns in major and minor tonalities and songs without words in a variety
of tonalities and meters. The remaining two groups were identiﬁed as the control
group and received 30 weeks of instruction using traditional rote songs with
words in major and harmonic minor tonalities and duple and triple meters.

The experimental and control groups were administered Primary
Measures of Music Audiation at the end of October and then again at the
beginning of May. It is unclear whether the Tonal subtest alone was
administered to the children or whether both subtests were administered. Kane

does report that only the data from the Tonal subtest was used in the analysis.

34

Kane provided instruction (the amount of instruction is not stated) to the
four classes during the months of October and November and used this time to
prepare two kindergarten classroom teachers to assume the role of music
instructor. She demonstrated guidance techniques with tonal patterns and song
instruction that were relevant to the experimental group and modeled techniques
for rote song instruction that were pertinent to the control group. Beginning in
December and continuing through April, the two kindergarten classroom teachers
provided 10 minutes of music instruction four days per week. Each teacher, one
male and one female, taught an experimental group and a control group. Kane
also conducted weekly meetings with the teachers to present new tonal patterns
and songs to be used the following week and to answer questions that arose
during the instructional period.

To determine the effect of singing instruction on developmental tonal
aptitudes, Kane examined the difference between the mean gain scores of
PMMA Tonal subtest for the experimental group and control group. Although not
statistically signiﬁcant, the experimental group had a higher observed mean gain
score from pretest to posttest than did the control group. Kane suggests this
may be due to the small number of subjects participating in the study and/or that
the beneﬁts of instruction may appear later when the child’s music aptitude
stabilizes. Kane concludes that there is no evidence to suggest that instruction
involving the singing of songs without words in a variety of tonalities and meters
has a greater effect on the developmental tonal aptitude of kindergarten children

than traditional instruction of rote songs with words.

35

DiBlassio (1985) examined four methods of tonal pattern instruction and
four methods of rhythm pattern Instruction to determine which had the greatest
effect on developmental tonal and rhythm aptitude, respectively. Sixteen classes
of ﬁrst-grade students participated in the study. The classes were divided into
two groups of eight. One group was administered the Tonal subtest of Primary
Measures of Music Audiation and received tonal instruction for 12 weeks. The
tonal pattern instruction was divided into the following four combinations of
tonalities: Group I, major and minor; Group II, major, minor, dorian, and
mixolydian; Group III, major, minor, and atonal; and Group IV, pentatonic. The
other group of eight classes was administered the Rhythm subtest of Primary
Measures of Music Audiation and received rhythm pattern instruction for 12
weeks. The rhythm pattern instruction was divided into the following four
combinations of meters: Group I, duple and triple; Group II, duple, triple, and
combined; Group III, duple, triple, unusual paired, and unusual unpaired; and
Group IV, duple, triple, combined, unusual paired, and unusual unpaired. At the
conclusion of the 12-week instructional period, students were administered the
Tonal or Rhythm subtest of PMMA, depending on their group classiﬁcation, to
determine the effect of instruction on developmental tonal aptitude or
developmental rhythm aptitude. No signiﬁcance was found according to
treatment. However, as expected, DiBlassio found that students with low
aptitude made greater gains than the students with high aptitude in both

treatment groups.

36

Holahan (1983) investigated “same" and “different” instructional
techniques on kindergarten students’ developmental music aptitudes. The
students were divided into the following fourlgroups: (1) teacher labeling tonal
patterns as the same, (2) teacher labeling tonal patterns as different, (3) teacher
labeling tonal patterns as the same and different, and (4) no labeling of patterns.
Prior to instruction, the students were administered the Tonal and Rhythm
subtests of Primary Measures of Music Audiation. PMMA was administered
again following the eight-week tonal pattern instruction period. Holahan found no
signiﬁcant difference among the children in the four treatment groups. However,
the developmental tonal aptitude of all students increased signiﬁcantly from the
beginning to the end of the instruction period. Holahan suggests that future
research may include a longitudinal study to determine if the results are due to
maturation or instruction.

Jessup (1984) attempted to determine the comparative effects of teacher
behavior on developmental music aptitude, music listening achievement, and the
performance of songs of second-grade students. PMMA was administered to
students as a pretest measure. Instruction lasted for 12 weeks, with two
teachers teaching two classes each. One class of each teacher received
primarily directive teacher instruction and the other received primarily indirect
teacher instruction. The teaching curriculum was selected from Book 2 of the
Silver Burdett music series. After 12 weeks of instruction, PMMA was
administered as a posttest. Students were also asked to complete items from

the Silver Burdett Music Competency Tests, perform the tonal and rhythm

37

patterns, and perform a rote song. Not surprisingly, Jessup found that the PMMA
scores of students with low music aptitude increased signiﬁcantly more than
those of high aptitude students. Also high aptitude students achieved
signiﬁcantly more on rote singing, pattern performance, and music listening tests
than low music aptitude students.

Rutkowski (1986) investigated the discrepancies found between the pitch
range of songs in music curriculum textbooks and the natural voice range of
kindergarten children (N = 162). The thought behind her research was that
singing beyond one’s vocal range may not only affect the use of singing voice; it
may also have an effect on developing music aptitude. In addition, Rutkowski
examined the possible relationship between children’s use of singing voice and
developmental music aptitude.

Over a period of 15 weeks, one group of kindergarten children received
music insthction that included songs with a restricted vocal range. Another
group of kindergarten children received music instruction that incorporated
textbook songs in their original form. Prior to the treatment period, the children
were taught a criterion song and ﬁve tonal patterns. Rutkowski’s Singing Voice
Development Measure (SVDM) was used as a pretest measure of children’s use
of singing voice. Reliabilities were as follows: (1) r = .74 for the pattern pretest;
(2) r = .93 8 .94 for the song pretest; (3) r = .91 8 .97 for the pattern posttest; and
(4) r = .88 8 .92 for the song posttest. She also administered the Tonal and

Rhythm subtests of Primary Measures of Music Audiation (Gordon, 1979) to

38

measure the children’s developmental music aptitude levels before treatment.
Both measurement instruments were also used as posttest measures.

The results of Rutkowski’s study suggest that music instruction using
songs within a restricted voice range does not signiﬁcantly affect developmental
music aptitude differently from instruction that includes songs outside of
children’s natural voice range. However, PMMA gain scores (Tonal and Rhythm
subtests) for the restricted voice range group were higher than those of the
control group. Furthermore, it appears that there may be a weak relationship (r =
.25 to .45) between children’s use of singing voice and their developmental tonal
aptitudes. The relationship between tonal aptitude and singing voice
achievement appears to be stronger than the relationship between rhythm
aptitude and singing voice achievement.

Two aspects of the 1996 study by Rutkowski were to determine (1) the
effect of individual and small-group instruction on developmental music aptitude
and (2) whether or not a relationship existed between children’s use of singing
voice and developmental tonal aptitude. The yearlong study included
kindergarten children (N = 99) from intact classrooms that were randomly
assigned to either the control (group activities only) or treatment group (small-
group and individual activities).

The Tonal subtest of Primary Measures of Music Audiation (PMMA) was
administered as a pretest, midtest, and posttest. The children’s song
performances were taped at the beginning and at the end of the school year.

Two judges using the Singing Voice Development Measure (SVDM) rated the

39

taped performances. The SVDM lnterjudge reliability was .90 for the pretest and
.99 for the posttest.

The results from Rutkowski’s study suggest that individual versus group
instruction does not affect children’s developmental tonal aptitude. Furthermore,
it appears that there is little correlation (correlations ranged from .135 to .207)
between children’s use of singing voice and developmental tonal aptitude.

Although the researchers above did not ﬁnd clear differences in aptitude
as a result of instruction, one cannot be sure whether this is a result of the choice
of treatment, maturation, or the nature of music aptitude. Other researchers
have found that instruction has an inﬂuence on children’s developmental music
aptitude, which leads one to believe that the nature of music aptitude is not the
reason for the lack of signiﬁcant differences.

Flohr (1981) examined the effect of short-term music instruction on ﬁve-
yeanold children’s (N = 29) developmental music aptitude. The Primary
Measures of Music Audiation was used as a pretest and posttest to assess the
effects of instruction.

The children were randomly divided in one of three groups. Music-l and
Music-ll groups received 25 minutes of instruction for a period of 12 weeks. The
instructional focus for Music-l was improvisation - question and answer games,
improvising to a bordun, improvising extensions to phrases, and improvising to
verbal stimuli. Music-II received the typical classroom experiences, while Music-

III received no music instruction.

4O

After 12 weeks of music instruction, Children’s PMMA scores in both
treatment groups were significantly higher from pretest to posttest than the
scores of children in the control group. This study suggests that children’s
developmental music aptitudes may be inﬂuenced by instruction. Due to a small
sample size, Flohr was unable to draw conclusions about the effectiveness
between the two types of music instruction.

In her developmental music aptitude study, Taggart (1997) investigated
the effect of age-appropriate instruction on the developmental tonal and rhythm
aptitudes of three- to eight-year-old children. Furthermore, she sought to
determine whether this effect varied depending on the children's age and
whether it would atrophy after instruction stopped.

One hundred eighty-three preschool through second grade children in at-
risk schools in Lansing, Michigan comprised the sample population. The
instructional period consisted of two 20—minute music classes every week taught
by a music specialist for an entire school year. Children had had minimal music
instruction as part of their school curriculum prior to this study. The treatment
consisted of singing and chanting in a variety of tonalities and meters, tonal and
rhythm pattern instruction in major and minor tonalities and duple and triple
meters, and movement instruction that focused on continuous, ﬂuid movement as
well as on beat movement.

The developmental music aptitudes of the preschool children were
measured pre- and post-instruction. The developmental music aptitudes of the

kindergarten, ﬁrst grade, and second grade children were measured prior to

41

instruction, immediately after instruction, and once again after a summer of
having no instruction. The Tonal and Rhythm subtests of Audie (Gordon, 1989b)
were administered to all preschool children. The Primary Measures of Music
Audiation was administered to all children in grades K-3. Pre-instruction and
post-instruction scores were compared as a measure of the children’s aptitudes
prior to and after music instruction. Also, in order to control for age, post-
instruction music aptitude scores after a summer of no instruction were
compared to pre-instruction aptitude scores of the children at the same grade
level. Because a different measure (Audie) was used for measuring the
aptitudes of preschool children than of children in grades K-3, the preschool
children’s scores were not considered in the comparison of differences in music
aptitude between grade levels.

With the exception of the preschool children’s tonal aptitude, aptitude
scores after instruction were significantly higher than aptitude scores before
instruction for all children, with the differences being greater the younger the
children were. Developmental music aptitude scores continued to increase over
the summer for children in kindergarten through grade three after a summer of no
instruction. In addition, at the beginning of the next academic year, the aptitude
scores of these children were signiﬁcantly higher than the aptitude scores of the
children the year before at the same grade level that had not received music
instruction. In other words, the increase in developmental music aptitude during
the summer for the children who had received music instruction the previous year

was greater than the increase that could be accounted for by maturation alone.

42

Taggart concluded the following: (1) appropriate music instruction
signiﬁcantly supports the developmental music aptitudes of three- to eight-year-
old children; (2) the effect of this instruction varies according to the age of the
child - the younger the child is the stronger the effect of instruction; and (3) the
effect of appropriate music instruction on developmental music aptitude holds
over time. Providing appropriate music instruction to children as early in their
lives as possible will contribute greatly to their musical potential.

A study of developmental music aptitude was conducted by Gordon
(1980a) to acquire information about the comparative music aptitudes of groups
of students who were of different environmental backgrounds and to further
substantiate the concept of developmental music aptitude.

In the fall of 1978, Primary Measures of Music Audiation was administered
to 167 kindergarten through third grade students enrolled in an inner-city school
in Rochester, New York. The children in kindergarten received music instruction
from their classroom teachers throughout the school year. First through third
grade students received two periods of music instruction each week by a music
specialist during the same academic year. At the end of the instructional period,
two Classrooms from each grade level were randomly chosen to take the Tonal
and Rhythm subtests of Primary Measures of Music Audiation.

The means, standard deviations, and split-halves reliability coefﬁcients for
the children in the inner-city school were compared to the corresponding data
from the standardization group as reported in the test manual. The children in

the standardization group received significantly higher scores on the Tonal

43

subtest than did the Children from the Inner-city school. Statistical signiﬁcance
was found for every grade level. The only signiﬁcant difference for the Rhythm
subtest was found at the third grade level, with the inner-City school students
receiving the higher test scores.

Gordon states that the inner-city school children may have received lower
tonal aptitude scores because of their limited informal and formal music
experiences. This statement is supported by the data showing a steady decline
of Tonal subtest scores for the inner-city school children in the ﬁrst, second, and
third grades and a steady increase of the Rhythm subtest scores of the children
in the ﬁrst grade to the third grade. In light of the results, Gordon concludes that
the concept and function of developmental music aptitude is objectively
supported and the quality of music instruction is of paramount importance.

Gordon (1980b) conducted another study of developmental music
aptitude in West Irondequoit, New York. In April of 1978, Primary Measures of
Music Audiation was administered to kindergarten (n = 127), ﬁrst-grade (n =
202), second-grade (n = 280), and third-grade (n = 264) students enrolled in one
of nine elementary schools in the city. PMMA was administered again
approximately two weeks later to test for practice effects from taking the test.
The differences between test and retest means were insigniﬁcant and the
standard deviations were comparable. Combined with their high test-retest
reliabilities (ranging from .60 to .76), the ﬁndings suggest, according to Gordon
(1986) that “even for tests which yield results that are expected to ﬂuctuate,

practice effects in taking the tests are negligible and should pose no threat to the

44

validity of Primary Measures of Music Audiation or Intermediate Measures of
Music Audiation” (p. 109).

The test results were then given to the music teachers so that they could
become familiar with the scores during the summer before instruction began the
following fall. The teachers used the suggestions provided in Part Six of the
PMMA Manual for teaching to individual children’s musical strengths and
weaknesses as indicated by aptitude test scores. The music teachers then
explained the test results to parents, and suggestions were offered to them for
providing their children with appropriate informal home experiences and formal
music instruction. Furthermore, teachers recommended that parents of children
who received composite percentile ranks of 80 and above provide some kind of
out-of—school special music instruction that would supplement the music
instruction offered to them by their regular school program.

In April of 1979, after the children had received approximately eight
months of individualized music instruction for one or two music periods each
week, PMMA was administered to them for a third time. Five hundred twenty-
three Children were still enrolled in the schools for the third administration of
PMMA, and these scores where the ones used in the analysis. One year of
physical and mental maturation was held constant, so that the differences in
PMMA tonal, rhythm, and composite mean scores between the two groups could
be attributed to the effect of instruction; the children in the control group had
received traditional music instruction that was provided by teachers who had no

speciﬁc knowledge of the children’s developmental music aptitude test scores,

45

whereas the children in the experimental group had received music instruction by
music specialists that was individualized based on tonal and rhythm aptitude
scores. Correlations were also computed for the 1978 and 1979 PMMA scores
of the children in the longitudinal group.

The results indicated that the majority of observed mean differences
between the experimental and control groups of children were significant, with all
differences favoring the experimental group. In addition, the differences tended
to be greater for younger students and more extreme when associated with the
Rhythm than with the Tonal subtest of PMMA. Gordon (1980b) acknowledges
that teachers had concentrated their lesson plans more on the rhythmic activities
than tonal activities. Correlations were calculated for the children’s 1978 and
1979 PMMA scores. The reliabilities of the children in the experimental group
were found to be considerably lower than the test-retest reliabilities computed in
April of 1978. Gordon (1980b) notes that it was expected as a result of
ﬂuctuating music aptitude due to intensive exposure and instruction over an
eight-month period. ‘

The results of this study suggest the following conclusions: (1)
developmental music aptitude is a reasonable concept, because scores ﬂuctuate
depending on the quality of the music instruction provided; (2) children’s
developmental music aptitude will develop more positively when the teacher has
knowledge of the music aptitude test scores; (3) the younger the children are

when music instruction is adapted to their individual music needs, the more they

46

will proﬁt from it; and, (4) Primary Measures of Music Audiation is sensitive to

developmental music aptitudes.

Summary

Many of these studies propose that music instruction inﬂuences children’s
developmental tonal aptitudes. Flohr (1981), Taggart (1997), and Gordon,
(1980a 8 1980b) and possibly Jordan-DeCarbo (1982) and Holahan (1983),
found that children received higher developmental tonal aptitude scores after
singing instruction. Furthermore, the younger children are when involved in
appropriate music instruction, the greater the gains in developmental tonal
aptitude (Taggart, 1997; Gordon, 1980b). Although Kane (1994) and Rutkowski
(1986 8 1996) found that developmental tonal aptitude was unaffected by vocal
instruction, this could simply be the result of an ineffective treatment. Levinowitz
(1987), Lange (2000), DiBlassio (1985), and Jessup (1984) ascertained that the
developmental tonal aptitude of children with low tonal aptitude increased more
than those children with high tonal aptitude, only some of which can be explained
by regression toward the mean. These studies give credence to the theory of
developmental tonal aptitude and to the supposition that appropriate instruction

increases developmental tonal aptitude.

The Effect of Instruction on Tonal Achievement

A multitude of researchers have studied children’s singing abilities. Some

have investigated the types of problems that are typical of children labeled as

47

inaccurate singers (Bentley, 1968; Davies 8 Roberts, 1975; Gould, 1969; Joyner,
1969). Others describe the effect of age on children’s singing ability (Boardman,
1968; Brown, 1988; Clegg, 1966; Cooper, 1995; Flowers 8 Dunne-Sousa, 1988;
Geringer, 1983; Goetze, 1986; Petzold, 1963; Sims, Moore, 8 Kuhn, 1982;
Moore, 1994; Moore, Brotons, Fyk, 8 Castillo, 1997; Sinor, 1985; Smale, 1988;
Yarbrough, 1991,1992) and the effect of gender on children’s singing ability
(Apfelstadt, 1984; Brown, 1988; Clegg, 1966; Cooper, 1995; Goetze, 1986;
Green, 1987, 1994; Jordan-DeCarbo, 1982; Joyner, 1969; Moore, 1994; Moore,
Brotons, Fyk, 8 Castillo, 1997; Petzold, 1963; Pederson 8 Pederson, 1970;
Phillips 8 Aitchison, 1997b; Rutkowski, 1986; Sinor, 1985; Smale, 1988; Stauffer,
1986; Welch, Sergeant, 8 White, 1995/1996). Still others examine the processes
involved in accurate singing: (1) pitch monitoring (Bentley, 1969; Davies 8
Yelland, 1977; Goetze, 1986; Smith, 1973; Welch, 1985, 1986), (2) pitch
discrimination (Apfelstadt, 1984; Bentley, 1968; Feierabend, 1984; Geringer,
1983; Pederson 8 Pederson, 1970; Smith, 1973; Stauffer, 1986; Zwissler, 1972),
(3) vocal range (Brown, 1988; Collins, 1981; Flowers 8 Dunne-Sousa, 1990;
Geringer, Nelson, 8 Kostka, 1981; Jersild, 1932; Moore, 1991; Vaughn, 1981;
Welch, Sergeant, 8 White, 1995/1996; Welch, 1979; Wilson, 1971), and (4)
group and individual singing (Joyner, 1971; Smith, 1973; Goetze, 1986; Goetze 8
Horii, 1989; Clayton, 1986; Smale, 1988; Green, 1994; Cooper, 1995). The
branch of research involving children’s singing that is most closely related to this
study are those studies that have examined the effects of different types of

instruction on children’s singing accuracy.

48

Characteristics of Singing Tasks

Some of the pitches, intervals, and melodic directions involved in singing
are more difﬁcult to perform than others. Several researchers have studied the
types of Singing tasks that may assist or impede Children’s vocal achievement.
The discussion in this section is limited to studies that focus on vocal instruction
that occurs over an extended period of time, as that is the focus of the current
study. Studies that involve a single session followed by the testing of children's
singing will not be included.

Jones (1971) hypothesized that a vertically arranged keyboard instrument,
manipulated by children, would be an effective means of improving the aural-
vocal performance of uncertain singers. Eight children in the second and third
grade (two boys and two girls from each grade level) were selected to serve as
subjects. The music teachers recommended a number of children based on their
having average music abilities and difﬁculties with singing. The recommended
students were then given a screening test consisting of twenty pitches. Children
considered for the study were the ones who correctly matched fewer than ﬁve
phches.

A test of aural-vocal skills was administered individually to each child prior
to instruction and once again at the end of instruction. The aural tasks included
the identiﬁcation of the number of pitches heard, identiﬁcation of whether pitches
were the same or different, and identification of the higher of two pitches. The
vocal tasks consisted of matching a single tone and matching tonal patterns.

The vocal portions of the test were tape-recorded and the aural portions were

49

notated in written form. The children’s performances on the pre-test were used
as a means of classifying their skill level according to a set of predetermined
levels of ability. This information formed the foundation of a practice plan for
individual Children to help them achieve the next level of ability. The researcher
kept a cumulative record for each child, noting the progress of the daily sessions.
Each child received a total of four hours of private music instruction. The
instruction was divided into sixteen 15-minute sessions throughout a four-week
period. The sessions were designed to follow a set of sequential steps that
moved the children from one level of singing ability to another.

The Magnus electric reed organ was used for developing vocal skills. The
pitches falling within the range of children’s voices were labeled with various
colors and corresponding pitch names, so children without keyboard experience
could manipulate the instrument. A light was installed on the bottom of the
keyboard to indicate to the children that their singing was accurate.

Each instructional period was designed to follow sequential steps in the
levels of singing ability. The steps consisted of the following: (1) distinguishing
the difference between extreme high and low pitches through the spatial
relationships that were kinesthetically experienced on the keyboard; (2)
distinguishing the difference between high and low pitches within one octave; (3)
matching a single pitch; and (4) matching two or three consecutive tones. After
singing with the instrument, the children sang after the instrument and were then

asked to state if their singing matched. The light attached to the bottom of the

50

keyboard was used at this point in the sequence to inform the children whether
their singing was accurate.

Songs were taught to the children as soon as they could sing a series of
pitches. Three pitch songs incorporating intervals of a second were introduced
ﬁrst. The song range gradually expanded to include intervals of a minor third,
major third, and a fourth. Near the end of the study, children were asked to play
the starting pitch of a familiar song using the "silent keyboard.“ As the children
sang the song, they were asked to follow the notes on the keyboard.

The results directly related to vocal achievement show that there were
gains in the children's pitch matching scores from pretest to posttest. It is not
indicated whether these gains were statistically signiﬁcant. Jones concludes that
having children play isolated pitches or patterns on the vertical keyboard prior to
singing intensifies concentration on the sound or sounds and, in turn, increases
vocal accuracy. In addition, Jones found that children were more successful
Singing melodic patterns and songs than single tones or scales. This is important
to consider when designing singing tasks that involve young children. Since
children's singing may be more accurate when performing songs rather than
single pitches or scales, the post-instructional testing procedures for the present
study will include song performance.

Jarjisian (1983) examined the effects of singing instruction using
pentatonic and diatonic patterns on the rote singing achievement of young

children. The participants in this study were three intact ﬁrst grade classes in

51

each of two elementary schools in Philadelphia. The schools differed with
respect to socio-economic status, teachers, and amount of instructional time.

The children in one school received 45 minutes of music instruction once
per week during a four-month period. The other school received two 30-minute
periods of music instruction per week for four months. Each of the three classes
in both schools was designated as T1 (diatonic pitch pattern instruction), T2
(pentatonic pitch pattern instruction), or T3 (a combination of diatonic and
pentatonic pitch pattern instruction). Ten to fourteen minutes were dedicated to
pattern instruction each week. The remaining class time included singing rote
songs, rhythm activities, listening activities, and movement activities.

Four criterion songs (two diatonic and two pentatonic) were taught to the
Children during the last month of the treatment period. Jarjisian did not indicate
the tonality of the two diatonic songs. Each child performed the four songs
individually, without accompaniment. The performances were tape-recorded and
then rated by two independent judges. The children’s level of singing
achievement was determined through a ﬁve-point continuous rating scale. The
interjudge reliabilities ranged from .72 to .86.

Analysis of the data revealed that the T3 group received signiﬁcantly
higher mean scores than did the T1 and T2 groups. There was no signiﬁcant
difference between the T1 and T2 groups. Children in ﬁrst grade receiving a
combination of diatonic and pentatonic pitch pattern instruction showed higher
levels of singing achievement than children who received only diatonic or

pentatonic pitch pattern instruction. Jarjisian concludes the following: (1) diatonic

52

pattern instruction aids the development of a sense of tonal center; (2) pentatonic
pattern instruction contributes to in-tune singing or a sense of melodic contour;
and (3) a combination of the two types of pattern instruction offers children the

best of both methods.

Text Versus No Text

Do children learn to sing more accurately when song instruction includes
text or when it does not? Gordon (1997c) states that the text serves as a
diversion from music Ieaming. When a song is taught using words, students
concentrate on the text of the song rather than on audiating tonal patterns,
rhythm patterns, tonality, and meter (Gordon, 1997a). Furthermore, inaccurate
singers may be inclined to use their speaking voice when words are used.
Learning songs with text may be difficult for young children, so Gordon
recommends that text be added only after children can accurately perform the
melody of a song. Andress (1986) also questions if the distraction of words is a
factor in the young child’s inability to accurately perform the melodicand rhythmic
aspects of a song.

Goetze (1986), Smale (1988), and Welch, Sergeant, 8 White,
(1995/1996), examined the effect of songs with text and songs without text on
children’s vocal accuracy. Goetze and Smale found no signiﬁcant difference in
children's pitch accuracy when singing a song with text or without text. Welch,
Sergeant, 8 White found that children were more accurate when singing songs

without text. Goetze, Smale, and Welch, Sergeant, 8 White did not investigate

53

the effect of song instruction on singing achievement and, therefore, will not be
discussed in depth in this discussion. The following studies (Richner, 1976;
Levinowitz, 1987; Lange, 2000; Kane, 1994) examine the effect of song
instruction with and without text on the tonal achievement of young children and,
therefore, will be included in the discussion.

Richner (1976) investigated the effects of various methods of music
instruction on the singing ability of inaccurate singers in grades three, four, and
ﬁve. Eight hundred forty-three students from four different elementary schools
within the same school district were screened to determine which students were
inaccurate singers. The initial selection of inaccurate singers was made based
on the student’s ability to sing familiar songs within a group. From the initial
screening process, 287 students were labeled as inaccurate singers. Students
from each grade level were randomly selected from the list of 287 students to
participate in the second part of the screening process. These students were
interviewed and asked to sing familiar songs individually. Eighty-two students
were found to be inaccurate singers in both group and individual contexts. Due
to a variety of reasons, the ﬁnal sample for the study consisted of 77 students (27
in third-grade; 23 in fourth-grade; 27 in ﬁfth-grade).

One of four treatments was randomly assigned to four different schools.
Students in the Treatment I group received no formal music instruction.
However, the classroom teachers did provide informal music experiences as
usual. Students in the Treatment ll group were taught by a music specialist and

received two 25-minute periods of music instruction per week for eight weeks.

54

These students remained with their intact classes for music instruction. The
inaccurate singers in this group did not receive individualized instruction.
Instruction for this group included rhythm activities, singing, instrumental
performance, note reading, listening, and creative activities.

Students in the Treatment Ill and Treatment IV groups were also taught by
a music specialist and received two 25-minute periods of music instruction per
week for eight weeks. The inaccurate singers in these two treatment groups also
received music instruction with accurate singers in groups of 10 selected from
each grade level. This was to insure that inaccurate singers were not placed in a
situation that could lead to poor self-concepts. The group size was kept to a
maximum of 10 in order to allow time for individual instruction.

Music instruction for Treatment lIl consisted entirely of singing songs with
text. The inaccurate singers sang only within the group; they were not asked to
sing individually and received no special attention. Music instruction for
Treatment IV included 20 minutes of remedial vocal training for the group and
individuals. Students were instructed on producing accurate single pitches within
the range necessary for a given song. When this was accomplished, students
were taught songs that contained stepwise patterns, followed by songs that
contained skipping patterns. All songs were sung without text. The remaining
ﬁve minutes of each class period was used to sing songs (with text) for
“enjoyment.”

Richner developed a 55-item “Pitch Reproduction Test” to measure the

student’s singing accuracy prior to and after instruction. Part I tested the

55

student’s ability to accurately reproduce one pitch at a time. Part II and Ill
measured the student’s ability to accurately sing melodic phrases. Both parts of
the test were divided into an A and B section. The A sections measured the pitch
accuracy while singing melodic phrases without text. The B sections measured
the pitch accuracy while singing melodic phrases with text. The strobotuner, an
electronic device that indicates degrees of pitch accuracy, was used in order to
“eliminate the need for subjective judgment about pitch accuracy” (p. 35).

An independent music specialist administered the test to each individual
child. Using the strobotuner, the test administrator rated the children’s singing
(individual pitches and individual pitches within each melodic phrase) as either a
match or a no-match. Richner reported reliability for the “Pitch Reproduction
Test” as .97.

Signiﬁcant differences did exist between the treatment groups at the ﬁfth-
grade level and at the third-grade level. There were no signiﬁcant differences
between treatment groups at the fourth-grade level. The pitch accuracy of ﬁfth-
grade inaccurate singers in the Treatment IV group Improved signiﬁcantly in
comparison with the other treatment groups. The pitch accuracy of third-grade
inaccurate singers in the Treatment Ill and IV groups improved signiﬁcantly in
comparison with students in the Treatment I group.

The students in the Treatment I group received signiﬁcantly higher mean
scores when singing melodic phrases without text than when singing with text. In
addition, when the total sample was considered, students received signiﬁcantly

higher mean scores when singing using a neutral syllable. Inaccurate singers

56

who received song instruction without text (Treatment IV) showed signiﬁcant
improvement in their ability to sing melodic phrases without text when compared
to inaccurate singers in Treatment l and Ill groups. Inaccurate singers who
received song instruction without text (Treatment IV) showed signiﬁcant
improvement in their ability to sing melodic phrases with text in comparison to all
other treatment groups. It is important to note that Treatment IV group received
individual attention, and, therefore, one cannot be sure if the results were due to
the effect of text/no text or to individualized instruction.

Levinowitz (1987) examined, in part, the effects of song instruction with
words and without words on the singing achievement of children in kindergarten
and ﬁrst grade. The children were randomly assigned to one of three treatment
groups: (1) songs primarily with words, (2) songs primarily without words, and (3)
songs always with words. She composed two criterion songs, one in major and
one in minor, so that the songs would be unfamiliar to the children. Levinowitz
created a tonal rating scale to rate the children on the use of singing voice,
accuracy of intonation, and ability to perform the resting tone in the context of a
song. The results of this study show no signiﬁcant difference in singing
achievement when instruction includes the use of songs with words or songs
without words.

Lange (2000) also investigated the effect of the use of text in music
instruction on the tonal accuracy of kindergarten students. Two classes (one

from each of 2 schools) were randomly selected to receive song instruction

57

primarily with text for a period of 24 weeks. During the same 24 weeks, the
remaining two Classes received song instruction primarily without text.

Lange selected a major/duple criterion song and a dorian/duple criterion
song from the folk literature. The tonal accuracy of the children’s performances
was rated using a ﬁve-point continuous researcher-designed rating scale.
Lange's results concur with those of Levinowitz (1987) that song instruction with
words verses song instruction without words made no difference in children’s
tonal accuracy when singing.

A portion of Kane’s study (1994), not discussed in the previous section,
examined the effects of tonal pattern and song instruction on the singing
achievement of kindergarten children. She was speciﬁcally interested in the
difference between instruction that included the singing of tonal patterns and
songs without words in a variety of tonalities and meters and instruction that
included the traditional teaching of rote songs with words in major and harmonic
minor and duple and triple meters.

At the conclusion of the instructional period, each child individually sang
two criterion songs, one in major tonality and triple meter and one inharmonic
minor tonality and duple meter. Both the experimental and control groups sang
the same songs, with the exception that the experimental group sang without
words and the control group sang with words. Both criterion songs were
performed twice, once with melodic accompaniment and singing by Kane and
once individually. A total of four performances were tape recorded and used for

analysis.

58

Two independent judges evaluated the children’s singing achievement
using a ﬁve-point continuous rating scale. lnterjudge reliabilities ranged from .79
to .83. Although not statistically signiﬁcant, the experimental group received a
slightly higher singing achievement mean score for both of the accompanied
songs and for both of the unaccompanied songs. Kane suggests this may be the
result of the classroom teachers singing songs with words to the children outside
of the “music class.” The control group had many more opportunities during the
day to sing songs with words, whereas the experimental group received only 10
minutes of song instruction without words each day. In addition, the criterion
measures are questionable, in that one group sang with text and one group sang
without text. It seems the discrepancy between singing tasks would render the
results meaningless. Therefore, the results of this study should be viewed with

cauﬁon.

Summagr
Three of these studies (Levinowitz, 1987; Lange, 2000; Kane, 1994) found

no difference in the tonal achievement of children resulting from whether
instruction included the use of text. Conversely, Richner (1976) found that vocal
accuracy increased as a result of the absence of text in instruction. Goetze,
Cooper, 8 Brown (1990) remark that “there is insufﬁcient evidence to conclude
whether children sing more accurately with or without text. The inconsistency of
ﬁndings suggests that more research is appropriate” (p. 32). While the present

study does not include an inquiry into the effect of text on children’s vocal

59

accuracy, it will incorporate the following: (1) all songs will be taught initially
without text throughout the treatment period; (2) text will be included after
children have learned to sing the melody and only when the text is a necessary
component of the song; (3) the criterion songs will be sung without text; and (4)
the Tonal Rating Scale (Levinowitz, 1987) will be used as a model to rate the
children’s song performances. Levinowitz reported high interjudge reliabilities (r
= .89 and .94) for the rating scale, and it most closely relates to the needs of the

present study.

Effect of Instruction on Improvisation

Improvising and composing are important means for learning and knowing
music (National Standards for Arts Education, 1994; Sherman, 1971). Both
improvisation and composition inform a teacher about a student’s level of musical
understanding. The type of instruction children receive may be a key factor in
developing the skills and cognitive understanding needed in order to improvise
and compose in meaningful ways. Although composition is an important
component of the music experience, it lies outside the realm of the present study
and, therefore, will not be included in the discussion. For the purpose of this
study of young children, this study of related literature will focus only on research
involving the effects of instruction on improvisation.

Joseph (1983) sought to determine if a kindergarten music program based

on Dalcroze Eurhythmics would develop children’s instrumental improvisational

6O

skills. She proposed that children receiving improvisation instruction would
outperform children who did not receive improvisation instruction.

Fifty-seven kindergarten children from two private schools served as
subjects. The design of the study consisted of two primary groups - children
receiving Eurhythmics instruction (experimental group) and those who did not
(control group). In addition, half of the children (Experimental Group I) received
Eurhythmics with chanting and instrumental improvisation instruction while the
other half (Experimental Group II) received Eurhythmics without chanting and
instrumental improvisation instruction. The instructional period was one school
year. Joseph taught two 25-minute music classes for each of the experimental
groups two times per week. A classroom teacher with a Bachelor’s Degree in
music taught the control group (Group III). The control group’s music classes
were not scheduled as a specific class time but instead were integrated into the
school day. Music activities for this group included singing with guitar
accompaniment, listening to records, and movement.

Joseph designed the Test of Kindergarten Child’s Aptitude for
Music/Movement Improvisation as a pretest/posttest measure. However,
unforeseen circumstances required that she change the design, and therefore
she used the test at the end of instruction only. The test was divided into three
sections. The ﬁrst section measured the child’s ability to discriminate between
and physically respond to same and different visual stimuli. In the second
section, children were asked to respond with large body movements to music to

determine if the children could coordinate movement with music. The tasks in

61

the third section involved the repeating and creating of patterns while maintaining
a steady tempo. The children were asked to perform the patterns by chanting,
clapping, and playing a drum. For the ﬁnal item in section three, children were
asked to improvise feely on a set of chromatic bells in order to determine if the
concept of pattern would be transferred to performance on the instrument. The
test was administered to 30 individual children (ten children randomly selected
from each group).

From the results of Joseph’s study (1983), it appears that children who
received Eurhythmics training were more adept at recognizing, associating
movement with, and responding to familiar patterns in unfamiliar music than
those children who did not receive such training. Furthermore, children receiving
Eurhythmics training with improvisation outperformed children not receiving
improvisation in their use of patterns when freely improvising on a set of
chromatic bells. It is also possible that Joseph's ﬁndings are the result of
instruction given by two different teachers rather than as a result of treatment.
Joseph taught the experimental group, whereas a musically trained classroom
teacher taught the control group. The results would have more meaning had
both teachers taught an experimental and control group.

Partchey (1974) examined the effects of three learning strategies on the
ability of sixth-grade students (N=86) to improvise melodies. Students were
randomly assigned to one of three groups: 1) Delayed feedback (n = 30) -
students immediately heard a replay of their improvisation before improvising the

second time; 2) Precomposed models (n = 28) - students heard a correct

62

response in expectation that it may positively inﬂuence subsequent
improvisations, and; 3) Repetition or control group (n = 28) - students received
the same number of opportunities to improvise but did not experience one of the
two experimental strategies. This group served as the control for the effect of
practice on the pretest/posttest scores.

Each student was pretested prior to instruction and then tested after
instruction. During the three-session instructional period, individual students
were given equal opportunities to improvise on an alto or tenor xylophone to pre-
recorded piano accompaniments. The tonalities of the accompaniments were
pentatonic, diatonic, minor blues, and whole tone - tonalities that contained few
leading tones. Accompaniment styles were jazz, rock, and blues. The
accompaniments were played ﬁrst for the student to listen to before improvising
the ﬁrst and third times; the second improvisation introduced the experimental
variable.

The pretest and posttest scores in this study were a measure of melodic
improvisational ability designed by Partchey speciﬁcally for the experiment.
Partchey’s two main criteria for judgment of the quality of the improvisations were
performance and creativity.

The primary performance criterion Partchey used was duration, having to
do with the subject’s ability to perform with the pulse of the stimulus and the
creation of clearly identifiable rhythm patterns. The two creativity criteria
Partchey used were variety and unity. The variety element of the improvisations

was examined in terms of tonal and rhythm contrast; the ability to generate these

63

features was taken as a measure of creativity. The unity element of the
improvisations was examined in terms of phrasing, repetition, imitation, and pitch
direction used by the student.

The criteria were numerically rated by three judges on a scale of 1
(weakest) to 9 (strongest). These scores were acquired for performance,
creativity, and the combination of performance-creativity. lnterjudge reliabilities
were determined to be .91 or above for each set of scores.

A repeated-measures ANOVA was conducted to determine if any
differences existed between the groups. There were no signiﬁcant
pretest/posttest mean score gains between the groups. Statistically signiﬁcant
pretest/posttest differences occurred between all groups in the creativity and
performance-creativity combined scores. In spite of these results, the Group 1
feedback strategy appeared to be slightly more effective at developing creativity
than the other two strategies. Partchey partially attributed this to the affective
responses generated by the students upon hearing their own improvisations
played back for them. Thirty-nine percent of the students did not beneﬁt from the
learning strategies - they experienced no gain or losses in the mean pretest-
posttest score. Partchey states that this may be due to factors of motivation,
interest, and awareness of the experimental procedures.

Overall, Partchey concluded that all three strategies do induce some
measure of Ieaming as manifested in the ability to improvise melodies. All three
strategies improved improvisational ability. Partchey also concludes that

creativity scores could improve independently of the performance skills.

64

Wig (1981) examined the viability of music composition strategies in
teaching sixth- and eighth-grade band students’ (N = 28) to improvise. In
addition, Wig examined the relationship between improvisation ability and two
other variables, performing ability and academic achievement. Students were
tested prior to and after instruction with The Watkins-Famum Performance Scale
and the Measure of Melodic Improvisation Ability. Students were administered
the SRA Achievement Series (a measure of academic achievement) by the
school district prior to instruction. The students received instruction in music
composition strategies for a seven-week period, one 50-minute session per
week. The compositional strategies included retrograde, inversion, retrograde
inversion, and embellishment of melodies. As part of his Measure of Melodic
Improvisation Ability, Wig asked subjects first to create a short song, about 10
seconds in length, and to think of two ways to vary it. They were to vary their
melodies in one of four ways taught to them in the composition lessons.

The improvisations were rated from one through ﬁve on four criteria -
creative use of pitch, creative use of intensity, creative use of duration, creative
use of form. lnterjudge reliabilities among the three judges were .80, and
intrajudge reliabilities for two judges were .92 and .94 for the composite scores.

Results showed that signiﬁcant gains were made in improvisation ratings
after the composition instruction. No relationship was found between the
improvisation gain scores and performing skill or between the improvisation gain

scores and academic achievement.

65

Amchin (1995) investigated the effects of teacher/student verbal
interactions on the improvisations of fourth- and ﬁfth-grade students (N = 129).
He sought to “measure whether exposure to improvisation activities found in the
Orff-Schulwerk along with teacher-student interactions designed to help children
evaluate their creative responses inﬂuences children to produce more musical
improvisations” (p. 8). Specific problems included measuring the signiﬁcant
differences 1) among mean scores on Webster’s Measure of Creative Thinking in
Music, Version 2 between children who had teacher-led discussions about their
improvisations and those children who did not; 2) among means scores on the
Measure of Instrumental Creative Musical Response (measure of improvisational
abilities) between the two groups of students; 3) among mean scores on
Intermediate Measures of Music Audiation between the two groups of students,
and; 4) in adjusted mean scores for the test of creative musical thinking and
improvisational abilities, adjusting using music aptitude mean scores as the
covariate.

Students from six intact classrooms from four different schools
participated in the study. The classes were randomly assigned to either the
control (n = 65) or experimental group (n = 64). The instructional period
consisted of 15-minute treatments twice a week for 23 weeks. Students in the
control and experimental groups were similar with respect to gender, Cognitive
Abilities Test scores, grade, musical background, interest in music, participation

in musical activities, and music aptitude.

66

Two activities occurred during each instructional session. The ﬁrst activity
consisted of a warm-up imitation game using melodic selections from the Orff-
Schulwerk. The selections used during the first seven weeks of instruction were
in duple meter. Melodies in triple meter were used during the next seven weeks
of the instructional period. Rote melodies in compound meter were used during
the third seven-week period. Selected melodies in duple and triple meter were
used during the remaining weeks of the study. The warm-up activities included
body percussion, vocal imitation, and beat accompaniment while learning each
selection. The students then performed the rote melodies on xylophones
following the teacher’s model.

The second activity in each of the instructional sessions was the melody-
completion task. An antecedent-consequence phrase was performed for the
students. Individual students were then asked to create their own consequent
phrase following the original model. Students in the experimental group were
then asked specific questions: “What is the difference between the model I gave
to you and your ending?,” “How is yours alike or different from the one I played
for you?," and “Can you think of another ending?” (p. 152). Finally, students
were given an opportunity to improvise a second ending. Questions were not
addressed to students in the control group prior to their second improvised
ending.

Webster’s Measure of Creative Thinking in Music, Version 2 (measure of
creative musical thinking) and Intermediate Measures of Music Audiation

(measure of music aptitude) were used as pretest/posttest measures. Froseth’s

67

How I Feel About Music (HIFAM) survey and the Cognitive Abilities Test (CAT)
were administered in the pretest phase of the study. The Post-Treatment
Survey-2 (PTS-2) and the Measure of Instrumental Creative Musical Response
(measure of improvisation ability) were administered during the posttest phase of
the study.

Results indicated no signiﬁcant difference in means scores on the MCTM-
ll, MICMR, and IMMA between the experimental and control groups. Also, there
were no signiﬁcant differences in MCTM-ll and MICMR scores after using IMMA
as a covariate.

Kalmer and Balasko (1987) examined the improvised songs of Hungarian
preschool children (N = 30) to determine if the music they were learning in their
classes created a musical vocabulary (referred to as “mother tongue”) that was
reﬂected in their improvisations. The children were from two nursery schools
(school A and school B) in Budapest, Hungary. Children were divided into three
groups as follows: (1) group A-m (n = 10), attended the school’s regular music
program and in addition participated in two additional 30-minute music periods of
“Special playful singing-music activities” (p.79) that was organized in a “very
systematic and expert way” (p. 79); (2) group A-c (n = 10), the control group,
participated in the regular nursery school music activities only, and (3) group B (n
= 10), was taught by a “teacher with a remarkably creative personality, where
music constituted an integral part of everyday classroom life” (p. 79).

All children were taught two counting rhymes (one with meaningful text

and one with a nonsense text) that could be recited in duple meter. Individual

68

Children were then asked to sing the rhymes as songs, beginning ﬁrst with the
nonsense text. The children created different melodies until they said they did
not want to did not want to create any more. The same procedure was repeated
for the meaningful text.

A total of 169 improvisations were recorded and analyzed. Improvisations
using familiar melodies were excluded from the analysis. All of the children were
able to create more than one melody. On the average, children improvised 5.6
melodies.

Results indicate that the children in group B signiﬁcantly outperformed the
other groups in nearly all areas. This group produced the most improvisations,
their melodies include the largest range of pitches, the largest intervals, and the
largest variety of tonalities. The children in group A-m were more careful in
performing their improvisations — “rules of the learned music material are
adhered to more stringently” (p. 80). Group A-m was the only group to improvise
complete melodies using the pentatonic scale. The poorest performance in the
application of learned rules and elements and creative productivity was by group
A-C.

The study was still in progress when the article was printed, and only
ﬂuency and originality had been examined as measures of creativity. Fluency
was measured in terms of the number of responses given by a child; group B had
far higher ﬂuency scores than the other two groups. Originality scores were
given for elements or features in the improvisations that were rare or novel in

nature. Group A-m’s mean originality score was ﬁve times higher than group A-

69

c’s mean originality score; group B’s mean score was three times higher than
that of group A-m.

Kalmer and Balasko suggest that the “musical mother tongue” (p.83) does
exist. Typical features of the Hungarian folk songs (volume, tonality, intervals,
structure, phrases, rhythm patterns) were identiﬁable in the children’s
improvisations. The authors concluded that “the most favorable condition for
fostering the children’s elementary music creativity is when the permanent
teacher of a group is musically well-trained and at the same time has a creative
personality” (p. 83). The results must be Viewed with caution due to the small

number of participants in the study.

Summary
Although Amchin (1995) found that instruction did not affect children’s

improvisations, Joseph (1983), Kalmer 8 Balasko (1987), Partchey (1974), and
Wig (1981) found that it did. As a result of instruction, young children can (1)
improvise tonal patterns (Joseph, 1983; Kalmer 8 Balasko, 1987), (2) improvise
music that incorporates musical structures (Kalmer 8 Balasko, 1987), and (3) can
maintain a steady beat and meter (Kalmer 8 Balasko, 1987). Different Ieaming
environments (Kalmer 8 Balasko, 1987), learning strategies (Partchey, 1974)
and composition instruction (Wig, 1981) can improve improvisational abilities.
Aside from the study of Kalmer and Balasko (1987), there is no research
on the affect of instruction on the vocal improvisations of young children. When a

child improvises vocally, one can be conﬁdent of the degree to which an

70

individual has internalized a music vocabulary, understands, and is able to
expresses intended musical ideas.

When a child improvises on an instrument, one cannot be completely sure
if the improvisation is a result of understanding or of luck in selecting the correct
pitches. More research is needed on the effects of instruction on children’s vocal
improvisations. Therefore, it is the intent of this researcher to incorporate this

element into the present study.

The Effect of Harmonic Accompaniment on Tonal Achievement

“Mainwaring (1951 ) feels that the primary concern when working with
children should be to awaken an interest in the aural effects of sounds heard
simultaneously and presented in musical combinations. The use of simple
polyphonic music can serve to develop concepts of consonance, dissonance,
and resolution which then serve as a basis for examining the qualities of chords
and chord progressions” (Petzold, 1966, p. 146). What is the role of harmony for
inexperienced young singers? Does the addition of harmony help or hinder the
development of singing achievement? Some studies have examined children’s
auditory perception of harmony (Berke, 2000; Bridges, 1966; Costa-Giomi, 1992;
Ritchie, 1962). The present study is interested in only those studies that have to
do with the effect of harmonic accompaniment on children's tonal achievement.

Sterling (1985) investigated the development of harmonic perception.
Sterling also examined the effects of three different harmonic contexts on

children’s simultaneous singing of familiar melodies. The harmonic contexts

71

were (1) melodic replication (no harmony — melody played one octave above
Child’s singing voice), (2) traditional tonal harmony (diatonic Chords with simple
progressions), (3) chromatic harmony (Chromatic, altered, extensions,
modulations), and (4) dissonant harmony (nondiatonic chords). The speciﬁcs of
Sterling’s study were to determine if differences existed in the vocal pitch
accuracy of children based on (1) grade level, (2) varied types of harmonic
accompaniment, and (3) dissimilar melodies (differences in pitch range and
rhythmic simplicity).

One hundred children in first, third, ﬁfth, and seventh grades (25 children
in each grade level) participated in the study. Children were randomly selected
from a pool of children that were preselected on the basis that they could
accurately sing (without accompaniment) “Jingle Bells,” “Twinkle, Twinkle Little
Star,” and “Old MacDonald.” Sterling wanted to use songs with which children
were already familiar, as there was no instructional component to her study.

Each child met with Sterling individually for testing. Data collection was
divided into two parts for each of the three test melodies. In Part I, children sang
the melody accompanied only by the melodic reference tape. They were allowed
at least two attempts to achieve a score of 35 as measured by the Pitch Master
but were limited to five attempts. The Pitch Master compares vocally-produced
pitches to those on a prerecorded reference tape. Any child that could not attain
the minimum score of 35 was eliminated from the study. Children who scored
higher then 35 on their ﬁrst effort were asked to sing again, and the highest of the

two scores was recorded. Part II consisted of children individually singing each

72

of the three familiar songs with the various types of accompaniment (12
performances per student).

All of the Children’s performances were recorded using the Pitch Master
device. The deviations from the model were measured, and the total number of
pitches that matched was counted and used to indicate the level of vocal pitch
accuracy. In order to control for the effect of position, Sterling alternated the
presentation order of the performances of melodies and harmonic
accompaniment.

The results of this study suggest that harmonic perception improves with
age. There was a significant difference in mean scores between children in ﬁrst
and third grades, with third-grade children receiving the higher scores. The
mean scores were not signiﬁcantly different for children in grades ﬁve, and
seven. However, children in seventh and ﬁfth grade scored Signiﬁcantly higher
than children in third grade.

The type of accompaniment had a significant effect on the children’s vocal
pitch accuracy. For each grade level, Children scored signiﬁcantly higher when
singing with the melodic reproduction and traditional accompaniments than when
singing with the chromatic and dissonant accompaniments. In addition, children
were singing with the melodic reproduction accompaniment scored signiﬁcantly
higher than when singing with the traditional accompaniment. There was no
signiﬁcant difference in scores between the chromatic and dissonant harmonic

accompaniments.

73

Finally, the nature of the melody had a signiﬁcant affect on the vocal
performance scores, regardless of the type of accompaniment. Children’s vocal
pitch accuracy scores were signiﬁcantly higher on “Old MacDonald” than for the
other two songs. “Twinkle” had lower mean scores than “Old MacDonald” and
was a little more difﬁcult for the Children to sing. The children received the lowest
mean score for “Jingle Bells.”

Although this study addresses a much-needed area of music research, the
results must be approached with caution. Sterling used children’s vocal accuracy
as a measure of their harmonic perception. This may be a problem, since there
is little correlation between children’s use of singing voice and developmental
tonal aptitude (Rutkowski, 1996). Sterling assumes that vocal accuracy is an
indication that children comprehend the sounds that they hear. She neglects to
address the fact that some children can sing accurately because they are good
imitators. Singing accuracy does not necessarily mean that one audiates well.
Using a valid aptitude test as an additional criterion measure may have been
useful.

The purpose of Kimble’s study (1984) was to consider the ability of
children to sing an added part. Speciﬁcally, Kimble compared the vocal accuracy
of children singing with a countermelody to their singing with a homorhythmic
part. Furthermore, the effect of order of presentation of the added parts was also
examined.

Seventy-six fourth-grade and 73 ﬁfth-grade students (N = 149) who met a

minimum criterion level for unison singing served as subjects. Students who met

74

the criteria were administered the Tonal Imagery subtest (melody and harmony)
of Music Aptitude Profile and the Otis-Lennon School Ability Test (IQ test). The
students were assigned to one of four treatment groups that were balanced in
terms of gender and singing ability. Two groups of students received melody
pre-training, and two groups of students received general song practice pre-
training during the ﬁrst four days of the study. All students received training in
singing the two added parts (countermelody and homorhythmic part) and were
tested on both. To control for order of presentation of the harmonic arrangement,
the treatments were counterbalanced across groups.

The daily treatment sessions were ﬁve minutes long and spanned over a
three-week period. During the ﬁrst four days of week two and week three,
students learned the added parts in a rote manner, unaccompanied, using the
syllable ”doo.” On the ﬁfth day of week two and week three, each student was
individually recorded singing the added part with a pre-recorded tape of children
singing the melody and the added part. Kimble reviewed the melody with the
students prior to the testing. The student was asked to sing the added part two
times for the recording. Three judges rated the recordings using a 7-point rating
scale of pitch accuracy. The higher of the two recording scores was used for
analysis.

Differences between the countermelody scores and the homorhythmic
scores were analyzed using a paired ttest. The homorhythmic scores were
found to be significantly higher than the countermelody scores. Kimble found

that melody pre-training made no difference in the pitch accuracy scores of

75

students. Students who did not receive pre-training received the same scores as
students who did receive pre-training. She also found that the presentation order
of added parts made no difference in student’s vocal pitch accuracy. In addition,
fourth- and ﬁfth-grade students were equal in their ability to accurately sing a
melody with an added part.

No signiﬁcant correlations were found between scores of pitch accuracy in
Singing a (1) countermelody and scores of aural discrimination, (2) a
homorhythmic part and scores of aural discrimination, (3) countermelody and
intelligence scores, and (4) a homorhythmic part and intelligence scores. Pitch
accuracy scores were correlated with (1) Melody MAP scores, (2) Harmony MAP
scores, and (3) school IQ scores. The countermelody scores were signiﬁcantly
correlated with Harmony MAP scores in three of the four instructional groups,
and no signiﬁcant correlations were found with the IQ test.

English (1985) investigated the effect of the amount of piano
accompaniment in beginning strings class instruction on the students’ intonation
and rhythmic stability. Although this study does not directly relate to children’s
singing, it does relate to children’s aural skills, their performance on an
instrument, and harmonic accompaniment.

Sixth-grade beginning strings students (N = 42) were randomly assigned
to one of three treatment groups, each receiving varying amounts of piano
accompaniment. The three treatment groups were as follows: (1) no piano
accompaniment; (2) 100% of instructional time with piano accompaniment; and

(3) 50% of instructional time with piano accompaniment.

76

In order to assess if equal amounts of musical ability existed between the
three treatment groups, English administered the Musical Aptitude Proﬁle (MAP).
English constructed a posttest measure, the String Performance Adjudication
Proﬁle (SPAP), for use in the evaluation of the children’s instrumental
performances. SPAP scores encompassed four areas: (1) group-prepared
individual intonation; (2) group-prepared individual rhythmic consistency and
accuracy; (3) individual sight-reading intonation; and (4) individual sight-reading
rhythmic consistency and accuracy. Each individual child’s instrumental
performance was taped and then rated by three judges.

The results for the pretest showed that the group receiving no piano
accompaniment scored signiﬁcantly higher on MAP prior to treatment than the
other two treatment groups. This indicates that the music aptitudes of the three
groups were unequal at the start of the treatment period, and therefore one group
may have had an advantage when achievement testing occurred. It is
unfortunate that the groups could not be divided so that each group had equal
levels of music aptitude between them. Had this happened, the results of the
study would have more meaning.

The no piano accompaniment group scored signiﬁcantly higher on each of
the four posttest variables than the other two treatment groups. The group of
students who had piano accompaniment 100 percent of the instructional time had
the lowest MAP scores and the lowest performance achievement scores on three

of the posttest variables. The one exception was the variable of prepared

77

rhythmic consistency and accuracy, for which the piano accompaniment groups
received relatively the same score.

Unfortunately, because there were signiﬁcant differences between groups
prior to treatment, the results of this study must be viewed with caution. As
expected, the group of children with the highest music aptitudes was also the
group of children who had signiﬁcantly higher performance ratings than the other
two groups. English neglected to do analysis of covariance to eliminate the
aptitude differences and, as a result, “It is unclear whether the group with no
piano accompaniment made higher posttest scores due to the treatment of
musical aptitude (this class made the highest score on the MAP pretest). It is
hoped that this study will provide impetus to future researchers to clarify these
issues” (p. 77).

Atterbury and Silcox (1993) examined the effect of piano harmonic
accompaniment on the singing ability of kindergarten children. The children were
divided into two groups. One group of children (n = 96) received all of their
singing instruction with piano accompaniment. The other group (n = 109)
received all of their singing instruction without piano accompaniment. The
researchers also examined the differences in singing ability that might occur
between low, average, and high music aptitude students, as well as the effect
that harmonic accompaniment might have on developmental music aptitude.

A pretest-posttest design with an additional posttest was used. A four
phrase, AABA song was taught to all of the children during the ﬁrst three weeks

of the school year. Each child was taped performing the song individually at the

78

end of the three-week period and then again during the ﬁrst week in June. The
children’s song performances were rated using a 4-point continuous rating scale
adapted from the scale developed by Rutkowski (1986). lnterjudge reliabilities
for the pretest and posttest were .747 and .859 respectively. In addition, the
Tonal and Rhythm subtests of Primary Measure of Music Audiation (PMMA)
(Gordon, 1979) were given to the children during the last weeks in May, and the
composite score was used in the analysis. The researchers attempted to
administer PMMA at the beginning of the school year and found that the children
were not developmentally ready to take such a test.

After one year of instruction, Atterbury and Silcox found no significant
difference between the treatment and control groups on song performance
scores. However, as would be expected, there was significant difference in song
performance between the three aptitude levels. Children with high
developmental music aptitudes received higher song performance mean scores
than children with average and low levels of aptitude. Also, the song
performance composite scores for children who had singing instruction with
piano accompaniment were approximately the same as children who received
singing instruction without piano accompaniment. The harmonic context did not
affect developmental music aptitude.

Atterbury and Silcox point out that the narrow range of their rating scale
may account for the lack of statistical differences in the analysis. They suggest
that this scale should be expanded for future use so that the measure is more

speciﬁc. They also suggest that a more accurate scoring system might be one

79

that focuses on speciﬁc phrases of a song instead of rating the entire song. They
conclude with a strong recommendation for future research to investigate the
effect of harmonic accompaniment on children’s singing.

Petzold (1966) was concerned with the relationships that may exist
between the accuracy with which children perceive and reproduce melodic
fragments when presented within varying harmonic contexts. He believed that
the key to accurate singing was the ability to perceive and retain the resting tone
of a tonality. Petzold also believed that “traditional” harmonic accompaniment
might serve as an invaluable aid to accurate singing. “Not only do the chords
themselves reﬂect the harmonic implications of the melodic line but they provide
the tonality, thus reinforcing the singers’ awareness of the tonal center of a given
song” (p. 147).

This discussion of Petzold’s study will be limited to his two research
questions that directly relate to the present study. The first question addresses
harmonic accompaniment and its effect on the accurate reproduction of a
melodic line. The second relates to the varying degrees of harmonic
accompaniment and their effects on Children’s singing.

Children (N = 360) in grades one through six participated in this study.
Ten melodic fragments from a 45-item test developed by Petzold were selected
based on item difﬁculty analysis - items selected were of medium and above
average difficulty. The selected items were used in the construction of the
Harmony Test, which measured Children’s vocal accuracy. The 10 melodic

fragments were presented within the context of three harmonic treatments. The

80

three types of harmonic accompaniment used were (1) a single chord that was
sustained throughout the entire melodic fragment, (2) primary Chords within
simple chord progressions, and (3) secondary chords, inversions, and moving
bass, alto and/or tenor lines within complex chord progressions that changed
with each pitch of the melodic fragment.

Three forms of the 40-item Harmony Test were then used to examine the
effect of harmonic accompaniment on the accurate reproduction of a melodic
line. The child’s task with Form A involved listening to the simultaneous
presentation of the melodic fragment and harmonic accompaniment (tape
recording with the piano as the performing medium) and then immediately
echoing that melody In conjunction with the harmonic accompaniment exclusive
of melody. Form B was analogous to Form A, with the exception that the child’s
response was unaccompanied. Form C was without accompaniment for both the
presentation of the melody and the child’s response. The sample was evenly
distributed among the three forms.

The accuracy with which children responded to a short melodic fragment
was not inﬂuenced by either a harmonic treatment of the melodic fragment or by
the harmonic treatment of the response. However, children did have signiﬁcantly
higher mean scores for accuracy when the presentation of the melodic fragment
and the response was supported by the primary chord treatment. The sustained
single chord was more helpful to children’s singing than the multi-chord

progression but not as helpful as the primary chords. There was not a signiﬁcant

81

difference between the three harmonic treatments when the accompaniment was
applied only to the presentation of the melodic fragment.

Petzold was interested in the effect of different types of harmonic
accompaniment on student's ability to sing melodic fragments accurately. The
present study is concerned with only one type of harmonic accompaniment and
the effect it has on children's ability to sing entire songs accurately and to
improvise within the context of major tonality. From the results of Petzold's
study, it seems reasonable to suggest that an uncomplicated harmonic
accompaniment is easier for Children to comprehend. However, is the
accompaniment using primary chords simple enough? Would Petzold have had
the same results using a root melody accompaniment instead of a sustained
single Chord? The sustained single chord gives little information about the
harmonic context of a melodic fragment or song. A root melody accompaniment
is simple in terms of the number of sounds presented at one time, yet it provides
a great deal of information about the tonality of the song.

The purpose of Azzara’s study (1992) was to enhance the music
achievement of elementary instrumental music students through a researcher-
designed improvisation curriculum. The speciﬁc problems of his investigation
were to examine (1) the effect of improvisation on music achievement and (2) the
effect of levels of aptitude on music achievement. Although Azzara’s study does
not focus speciﬁcally on harmony, the harmonic nature of the treatment is

relevant to the present study.

82

Fifth-grade wind and percussion music students from two schools served
as subjects. Students in both schools had completed one year of instrumental
music instruction using Jump Right In: The Instrumental Series (Grunow 8
Gordon, 1989). Azzara administered Music Aptitude Proﬁle (Gordon, 1965) to
students in the treatment and control groups prior to treatment. The
experimental treatment involved the use of an improvisation curriculum designed
for the study. Students were randomly assigned to the experimental treatment
by homogeneous instrument classes (three to six students per group). The
treatment period lasted 27 school calendar weeks. The treatment group (1)
audiated and improvised tonic and dominant function patterns in major tonality
using neutral and solfege syllables, (2) sang all tonic and dominant chord
members through a progression in major tonality, using neutral and solfege
syllables, (3) improvised through a tonic/dominant chord progression in major
tonality while other students sang the chord tones, (4) while others sang the
tonic/dominant chord progression in major tonality, individual students sang the
chord roots using neutral and solfege syllables (5) improvised duple meter
patterns, using neutral and rhythm syllables, (6) combined tonal and rhythm
improvisation, and (7) improvised to the chord progressions of various songs.
The control group did not receive improvisation instruction.

At the end of the treatment, all students performed three criterion etudes
written by Azzara to measure their music achievement. The ﬁrst etude was
student-prepared, the second etude was prepared with the help of the teacher,

and the third etude was sight-read. Four independent judges rated the etude

83

performances using researcher-designed ﬁve-point continuous tonal and rhythm
rating scales, and a researcher-designed ﬁve-point additive expression rating
scale.

Signiﬁcant differences existed between students who received instruction
that involved the use of the improvisation curriculum and those who did not. The
treatment group received higher ratings on the three criterion etudes than
students in the control group. As expected, signiﬁcant differences were also
found between aptitude levels. The group of students who were classiﬁed as
having high music aptitude demonstrated higher achievement. “With
improvisation skill, students can express their musical thoughts Spontaneously.
This expression is possible when students comprehend the tonal and rhythm
patterns and can combine and sequence them in a syntactical manner, they
internalize a sense of tonality and meter. This kind of understanding engenders
an understanding of harmonic progression. Harmonic progression is deﬁned in a
linear sense by (1) where the harmony changes (tonality) and (2) when the
harmony changes (meter)” (p. 69).

The students in Azzara's study had many opportunities to familiarize
themselves with harmonic progressions. They heard bass lines, sang bass lines,
part-sang the chords of harmonic progressions, and improvised melodies to the
bass lines of a variety of songs in major tonality. The children in the present
study will hear a variety of bass lines to rote songs and improvise tonic,
dominant, and subdominant functions in both major and harmonic minor

tonalities. The students in Azzara's study were much older and probably had the

84

readiness to sing bass lines and part-sing chords. These were not
developmentally appropriate tasks for the children in the present study and,
therefore, were not included. The control group in Azzara's study did not have
improvisation instruction. The present study will include improvisation for the
treatment as well as the control group. This may provide information about
whether hearing bass lines helps children develop a sense of ”where” and
”when” harmony changes.

In order to examine the effects of melodic and harmonic contexts on the
singing and aural discrimination abilities of children, Stauffer (1986) asked
children to echo melodic patterns within the context of different melodic and
harmonic accompaniments. Children in ﬁrst through third grade (N=310) served
as subjects. The children received twelve weeks of instruction, which included
melodic echo training tapes that incorporated various accompaniment contexts.
The four accompaniments were: (1) no context, (2) melodic context, (3) harmonic
context, (4) melodic and harmonic contexts. Intact classes within each grade
were randomly assigned to one of the four treatment groups.

Three data collection instruments were used as pretest-posttest
measures: Primary Measures of Music Audiation (PMMA), Melodic Echo Test
(MET), and Test of Singing Ability (TSA). The Wide Range Achievement Test —
Revised (VVRAT-R) was used as a pretest. It is an academic achievement test to
study the development of reading, spelling, and arithmetic. Stauffer used this
test to examine differences between the two schools and to determine if a

relationship existed between reading achievement and musical skills. The How I

85

Feel About Music survey, Froseth’s revised attitude and interest survey, was
used to collect additional data about the student’s musical background and
preference for musical activities.

Only the results that directly relate to this study will be reported. No
signiﬁcant differences were found between PMMA posttest mean scores for the
four main treatment groups. However, gains for the Tonal subtest of PMMA were
signiﬁcant for all children regardless of treatment group. Signiﬁcant differences
were identified in both Tonal pretest and posttest PMMA scores when children
were grouped by grade level, with consecutive grades scoring higher. In
addition, ﬁrst- and second-grade children showed greater improvement than
third-grade children. “It appears that improvement in aural discrimination skills
may have been more rapid for younger subjects” (pp. 129-130).

The MET contains twenty melodic patterns that students listen to and then
echo through singing. lnterjudge reliabilities for both the pretest and posttest
were .92. No signiﬁcant differences were found for MET between the four
treatment groups. Children’s ability to imitate melodic patterns was not affected
by the type of harmonic treatment they received. However, their ability to
accurately imitate these patterns did signiﬁcantly increase, regardless of
treatment. An interaction effect of melody, harmony, and grade was found.
Further analysis revealed that an interaction between melody and harmony was
evident for third-grade subjects only. In the ﬁrst and second grades, children in
the Melody treatment groups scored higher on MET than children in the No

Melody group. This indicates that the harmonic treatment may have been more

86

effective for children in the third grade than for children in the ﬁrst and second
grades. “These results may conﬁrm research, which indicates that harmonic
discrimination, develops at approximately age eight, or third grade” (p. 212).
“Some effect may be attributed to the interaction of melody and harmony for third
grade students, but it appears that the no harmony treatment may be more
beneﬁcial for ﬁrst-grade and second-grade subjects” (pp. 161-162).

The TSA measures children’s ability to accurately sing ﬁve rote songs.
The children’s performance of each song is rated using a five-point continuous
rating scale. The interjudge reliability for TSA was .82 for the pretest and .87 for
the posttest. Children in each grade improved in their ability to sing rote songs.
The mean TSA scores were not significantly different among the four treatment
groups after instruction. However, the posttest mean score for each group was
signiﬁcantly higher, regardless of treatment.

In conclusion, Stauffer (1986) suggests that (1) aural discrimination of
tonal patterns, vocal imitation of melodic patterns, and ability to sing rote songs
are strongly related, (2) aural discrimination and singing skills improve with
training regardless of musical context, (3) harmonic context used in music
training tasks may be more beneficial to older children than younger children,
and (4) musical skills are affected by maturation, with more rapid growth
occurring in earlier grades.

The treatment in Stauffer’s (1986) study differs from the present study in
that Stauffer used melodic echo training tapes to incorporate the various

accompaniments into instruction. The instructional procedures for the present

87

study will not include training tapes. The melodic accompaniments will be
provided through live performance on a variety of instruments, including the
voice. In addition, the treatment period for the present study will be twice as long
- 24 weeks instead of 12. The effects of a year of instruction are more likely to
hold up over time (Taggart, 1997) than the effects of a 12-week instructional
period (Flohr, 1981 ). The biggest difference between the two studies is the type
of harmonic accompaniment used during the treatment period. Stauffer (1986)
used no harmonic accompaniment, melodic accompaniment, harmonic
accompaniment, and melodic and harmonic accompaniment. It may be possible
that a simple harmonic accompaniment may increase children’s vocal accuracy,
whereas more complex harmonic accompaniments may not. Therefore, the
present study will use a root melody accompaniment in place of chordal
harmony. No studies to date have examined the effect of a root melody
accompaniment on children’s vocal achievement.

Hale (1977) investigated the effect of piano accompaniment on children’s
singing development. She speciﬁcally examined if kindergarten children’s (N =
247) singing should be accompanied on the piano (sometimes autoharp) using
(1) only harmonic accompaniment or (2) a three step sequence, beginning with
melodic reinforcement (melody played in octaves), followed by melodic
reinforcement and harmonic accompaniment together, and then ﬁnally harmonic
accompaniment alone.

The children were randomly assigned to one of the two groups on an

individual basis and not on the basis of membership in a particular class. The

88

treatment occurred throughout an entire school year. During a ﬁve-day period,
two criterion songs, one in major tonality and one in minor tonality, were taught to
the children. The harmonic accompaniments were used during instruction. At
the end of the treatment period, children were asked to sing two criterion songs
without accompaniment. Both criterion songs were in duple meter in order to
control for the effect of meter on the children’s song performance. Children sang
the two criterion songs again a week after the original performance to establish
test/retest reliabilities of the ratings.

Two independent judges using a ﬁve-point continuous performance rating
scale rated the children’s singing. When combining the ratings of the children’s
ﬁrst and second week song performances, the interjudge reliability was .71 for
the major tonality song and .71 for the minor tonality song. An interjudge
reliability (r = .83) for the composite score (major and minor combined) was also
reported. The judges’ combined ratings of children’s performance had
reliabilities of .76 for the major song, .77 for the minor song, and .92 for the major
and minor composite score.

Children who received song instruction with melodic and harmonic
accompaniment scored signiﬁcantly higher on the criterion measure for both the
major and minor song than did children who received song instruction with
harmonic accompaniment alone. As a result, Hale suggests, “piano
accompaniment retards the development of students’ singing ability” (p. 40).

There are several similarities between Hale’s study (1977) and the present

study. First, the children were asked to sing two criterion songs, one in major

89

tonality and one in minor tonality. Both studies used criterion songs in duple
meter to control for the effect of meter on the children’s song performance.

The Hale study differs from the present study on several accounts. The
children in Hale’s study were randomly assigned to one of the two groups,
whereas the present study used intact classes of children. Hale used only
children in kindergarten, while the present study incorporated children in
kindergarten and ﬁrst grade. The present study used a variety of instruments
when performing root melody accompaniments to control for the effect of
instrument type. Hale used piano and autoharp only. Another difference
between the two studies is that Hale used type of accompaniment (harmonic or
melodic/harmonic) as an independent variable, whereas the present study used
root melody accompaniment. Hale found that children who received song
instruction with melodic and harmonic accompaniment scored signiﬁcantly higher
those children who received song instruction with harmonic accompaniment only.
It may be possible that children in the melodic and harmonic accompaniment
group were merely imitating the melody as it was played. This may explain the
difference in song performance scores between the two groups. Therefore, the
present study used a root melody accompaniment instead of melodic

accompaniment in order to control for imitation.

Summam

There is inconclusive evidence regarding the effects of harmonic

accompaniment on children’s singing. Some studies found that harmonic

9O

accompaniment had no effect on children’s singing accuracy (Atterbury 8
Petzold, 1966; Stauffer, 1986). Other studies found that harmonic
accompaniment had a signiﬁcant effect on Children’s singing accuracy (Hale,
1977; Kimble, 1984; Sterling, 1985). In addition, no harmonic accompaniment
(English, 1985) or simple harmonic accompaniment (Kimble, 1984; Petzold,
1966; Sterling, 1985) appears to be more beneﬁcial than complex harmonic
accompaniments. Azzara (1992) found that students who learned to improvise
by hearing and singing bass lines and chords within the context of harmonic
progressions achieved at higher levels than students who did not have such
instruction.

Considering the studies reviewed in this section, the importance of the
present study becomes apparent. Research comparing the effects of simple
harmonic accompaniment with a cappella singing is in order (Goetze, Cooper, 8
Brown, 1990). Perhaps simpler is better. It is possible that even primary triads
are too much for young children to attend to. It is conceivable that root melodies
provide the most appropriate level of harmonic context for children who are
developing audiation skills. Therefore, the harmonic accompaniment used in the
present study will be the root melody or melodic line created by the fundamental
pitch of the harmonic functions found in the songs. Furthermore, in light of the
scarce number of studies involving the effects of harmonic accompaniment on
young Children’s vocal achievement and improvisation, this study is an important
step in the continuum of research that seeks to deﬁne factors that are beneﬁcial

to children’s developing musicianship.

91

Chapter Summary

The research reviewed in this chapter focuses on four major areas: ( 1) the
effect of instruction on developmental tonal aptitude; (2) the instructional factors
that affect children’s vocal accuracy; (3) the effect of instruction on improvisation;
and, (4) the effect of harmonic accompaniment and instruction on children’s vocal
accuracy. Studies that are Closely related to the aforementioned sections were
presented to provide the research base and the motivating force behind the
present study.

The researchers presented in this chapter concluded the following: (1)
developmental tonal aptitude may increase as a result of instruction; (2)
children’s singing accuracy may improve as a result of instruction; (3) children’s
instrumental improvisations can reﬂect a sense of tonality given instruction; and
(4) vocal instruction that includes harmonic accompaniment may or may not
improve children’s singing accuracy.

The studies in this chapter regarding the effect of instruction on
developmental tonal aptitude either compared the effects of different approaches
in vocal music instruction or compared the effects of instruction versus no music
instruction. The present study compares two different kinds of instruction —
instruction with and without harmonic accompaniment and its effect on
developmental tonal aptitude. All researchers discussed in this chapter used
Primary Measures of Music Audiation to measure children’s developmental

music aptitude. This study differs in that Intermediate Measures of Music

92

Audiation serves as a pretest and posttest measure, because Primary Measures
of Music Audiation was too easy for the children who participated in this study.

Most of the studies that focused on the effect of instruction on Children’s
tonal achievement had limited instructional time. The treatment periods ranged
from four hours to several weeks to a few months. The treatment period for the
present study spans more than seven months and includes more frequent
sessions. Rutkowski (1996) states that research with young children involving
Short treatment periods may result in premature conclusions. Additionally, in
many of the other studies, researchers entered into classrooms for the purpose
of conducting research and were not familiar with the children prior to the study.
This study varies from the others in that the researcher has taught music to all of
the children for two years prior to the study.

The studies on the effect of instruction on improvisation focused on either
a specific methodology (Dalcroze) or on instrumental improvisations. Only one of
the studies presented in this chapter sought to examine children’s vocal
improvisations. It is one thing to improvise using an instrument and quite another
to improvise vocally, which requires one to rely solely on audiation skills to
accomplish an improvisational task. When children improvise vocally, one has a
window into the level at which they have internalized a music vocabulary.
Therefore, unlike most studies involving young children and improvisation, the
present study incorporates vocal improvisation as a measure of children’s tonal

achievement.

93

The studies on the effect of harmonic accompaniment on tonal
achievement focused on either accompaniment versus no accompaniment or on
the difference between various harmonic contexts. The present research relates
most Closely to the harmonic/no harmonic accompaniment studies. The piano
served as the source of harmonic accompaniment for the studies discussed in
this chapter. The present study uses a variety of instruments, including piano
and voice. Furthermore, it appears that a simple harmonic accompaniment may
increase children’s vocal accuracy, whereas more complex harmonic
accompaniments may not. With this in mind, the present study uses a root
melody accompaniment in place of chordal harmony. No studies to date have
examined the effect of a root melody accompaniment on children’s tonal
achievement.

The present study seeks to fill some of the gaps in the studies presented
above by investigating the effect of (1) song instruction including root melody
accompaniment on developmental tonal aptitude, (2) song instruction including
root melody accompaniment on tonal achievement, and (3) song instruction
including root melody accompaniment on tonal improvisations. No other study to
date has examined the effect of harmonic accompaniment on improvisations. It
is the intent of this researcher to gain insight into the issues presented above and

to provide an impetus for further research.

94

CHAPTER THREE

DESIGN AND ANALYSIS

Sam Ie

Sixty-eight kindergarten students and 68 ﬁrst grade students (N=136)
participated in the study. The children in this study were from four intact
kindergarten classes (17 students per class) and four intact ﬁrst grade classes
(17 students per class). All of the students attended a private independent
school in Bloomﬁeld Hills, Michigan. The children in the school ranged from age
three through the ﬁfth grade and came from a primarily white, upper-class
background.

Children at the school maintained a six-day cycle. This meant that
teachers and children followed a schedule that was based on A through F Days
instead of the typical Monday through Friday schedule. For example, the ﬁrst
Monday of the school year followed an “A Day Schedule,” Tuesday followed a “B
Day Schedule,” Wednesday followed a “C Day Schedule,” Thursday followed a
“D Day Schedule,” and Friday followed an “E Day Schedule.” The schedule for
the second Monday of the school year followed the “F Day Schedule,” followed
by the “A Day Schedule” on Tuesday, the “B Day Schedule” on Wednesday, the
“C Day Schedule” on Thursday, and the “D Day Schedule” on Friday. The six-
day cycle continued in this manner for the remainder of the school year. The six-

day cycle was put into place at the school in order to create a sense of balance

95

within the school day. As a result of the schedule, children spent more time with
their homeroom teachers, had fewer interruptions in their day, and were able to
experience many special class offerings (music, drama, dance, art, physical
education, Spanish).

Children in kindergarten received 30 minutes of formal music instruction
twice per cycle. The children in ﬁrst grade received 40 minutes of formal music
instruction three times per cycle. This schedule continued for the duration of the
study.

The music curriculum that was typically used for all of the children in
kindergarten and two of the ﬁrst grade classes was based on Music Learning
Theory (Bolton et. al., 2001; Gordon, 1997; Taggart et. al., 2000), which is a
theory of how children learn music. Instruction that is based on Music Learning
Theory includes two parts: learning sequence activities and classroom activities.
Learning sequence activities, or pattern instruction, are research-based (Gordon,
1974, 1976) and sequentially introduce new skills, such as tonal and rhythm
solfege, identification of tonalities and meters, identiﬁcation of pattern functions
(i.e., tonic, dominant, macrobeat/microbeat, division/elongation), and
improvisation. Classroom activities include active music making through singing,
chanting, movement, and playing instruments. Children also apply to listening
what they have learned while making music. Classroom activities and learning
sequence activities are taught together and coordinated with one another in order
that children understand and comprehend the music they hear and perform. The

researcher taught these Classes. Two other music specialists taught the

96

remaining two ﬁrst-grade classes that were used in this study. The approach to
music instruction that was most familiar to one first-grade music teacher was
based primarily on Orff's philosophy. The underlying belief of Orff-Schulwerk is
that music is elemental, and instruction should be based on what children do
naturally. In general, learning is an integration of music, speech, movement,
drama, and performance experiences with Orff instruments. The approach to
music instruction that was most familiar to the other music teacher was that of
the Kodaly philosophy. This philosophy has grounding in folk song literature and
emphasizes music literacy — solfege, music reading through Curwen-Kodaly

hand signs, and writing.

Egan
The design of this study was treatment by levels. The treatment was
instruction including songs with no root melody accompaniment or song
instruction including root melody accompaniment. The level was grade in school

(kindergarten or ﬁrst).

Procedures
After the researcher’s guidance committee accepted the proposal, a copy
of the proposal was submitted to the University Committee on Research
Involving Human Subjects (UCRIHS). After permission for conducting the
research was granted by UCRIHS (see Appendix A), the researcher submitted a

copy of the proposal to the school administrators, received permission for

97

conducting research, and distributed information about the research study to
kindergarten and ﬁrst-grade parents (see Appendix B). A parental consent form
(see Appendix C) accompanied the letter. Participation in the study was entirely
voluntary, and a Child could withdraw from the study or refuse to respond at any
time without penalty. The researcher ensured conﬁdentiality of all student
responses. All of the consent forms were returned, and therefore all of the
children participated in the study.

This study involved a treatment group consisting of two randomly selected
kindergarten classes, one randomly selected first-grade class taught by the
researcher, and one of the two ﬁrst-grade Classes taught by one of the other two
teachers.

The researcher provided instruction to two kindergarten control groups,
two kindergarten treatment groups, one ﬁrst-grade control group, and one ﬁrst-
grade treatment group. Another teacher provided instruction to one ﬁrst-grade
control group and a different teacher instructed one first-grade treatment group

(Table 1).

98

Table 1

 

 

Teacher Assigments

Experimental Experimental Control Control
SK Researcher Researcher Researcher Researcher
1 Researcher Teacher 1 Researcher Teacher 2

 

For the treatment group, a root melody accompanied eighty percent of the
rote songs used in each instructional period. Other than the use of root melodies
for the treatment group, all children received the same instructional content. The
music instruction provided by all three music teachers included Learning
Sequence Activities (Gordon, 1992a, 1992b, 1992c), acculturation and
performance of rote songs and chants in an array of tonalities and meters,
vocally produced rhythm and tonal creativity/improvisation, expressive movement
activities, beat coordination activities, play parties, and instrumental performance
experiences using simple unpitched percussion. Song repertoire (see Appendix
D) and all lesson plans (see Appendix E 8 F) were developed by the researcher
and were adhered to by all teachers involved in this study. A 50-minute planning
meeting occurred on every C Day throughout the treatment period to insure that

instruction between the three music teachers was as similar as possible.

99

One factor that may affect the vocal accuracy of children is the model that
children hear. Studies suggest that the female voice and pitches that are within
the vocal register of the Child’s own voice are the best models for supporting
accurate singing (Brown, 1988; Clegg, 1966; Froehlich, 1977; Green, 1987 8
1990; Hermanson, 1972; Kramer, 1986; Moore, 1991; Petzold, 1966; Sims,
Moore, 8 Kuhn, 1982; Small 8 McCachern, 1983; Smith, 1963; Yarbrough, 1991,
1992). Because it is impossible for one teacher to sing both the song and root
melody at the same time, other options needed consideration. In order to control
for the effect of a tonal model, the simultaneous performance of the song and the
root melody was presented using a variety of performance mediums: bass
metalophone and xylophone bars; alto metalophone and xylophone bars; voice
recordings; live vocal performance; piano. In addition, when it was
developmentally appropriate, children in ﬁrst grade also heard and then sang the
root melody without the song and then with the song within the context of group
and individual Singing. Because it is not developmentally appropriate, the
children in kindergarten were not asked to perform the root melody in this way.
Part-singing is a difficult task for children this young (Hufstader, 1977;
Zimmerman, 1971).

The Intermediate Measures of Music Audiation was administered during
the week of September 11, 2001 and then again during the ﬁrst week in April
2002. Both the Tonal and Rhythm subtests were administered so that the

necessary information about individual’s strengths and weaknesses was

100

available for instructional purposes. However, only the Tonal subtest was used
in the analysis as a dependent variable.

Post-instructional performance testing occurred beginning the second
week of April 2002. Individual children were taken out of their regular class and
led into the music room. This room served as the music room for all but two of
the ﬁrst-grade Classes. These children were familiar with the room, because it
was where they had received music instruction for the last three years. In
addition, the researcher did not pose a threat to the children, because they were
all familiar with her as their past or current music teacher. After children gave
verbal assent (see Appendix G), they individually performed two rote songs
without text, one in major tonality/duple meter and one in minor tonality/duple
meter, and improvised an ending to a major tonality/duple meter song without
text (see Appendix H). All singing was performed without root melody
accompaniment.

Two criterion songs, one in major tonality and one in minor tonality, were
taught to all children four weeks prior to the end of the treatment period (see
Appendix I). The songs were composed by the researcher to insure that the
song content accurately reﬂected the criteria found in the rating scale. Both
criterion songs were in duple meter in order to control for the effect of meter on
song performance. In addition, both songs had an equal number of phrases.
Levinowitz, Barnes, Guerrini, Clement, D’April, 8 Morey (1998) found that the
reliabilities for their major criterion song were slightly higher than for their minor

criterion song. They suggest that this occurred because the minor criterion song

101

contained more phrases than the major criterion song, so children’s audiation
skills were challenged. The longer song may have left the Children with more of
a chance for inconsistency with their singing performance. The researchers
suggest that more research is needed before conclusions can be made regarding
Children’s performances of major versus minor songs. Finally, in light of the
research suggesting that children may Sing with greater accuracy when songs do
not include text (Richner,1976; Levinowitz, 1989), the criterions songs used in
this study were sung without text.

Solo singing was a common practice during regular music instruction, so
singing for the researcher did not seem out of the ordinary. All of these
performances were recorded using an Olympus Voice and Music DM-1 Digital
Voice Recorder so that they could be rated at a later time. Prior to recording, the
children heard the tonic and dominant pitch of the song they were about to sing
performed on a resonator bell. Then the researcher sang the appropriate
preparatory sequence (see Appendix J) before the children sang the criterion
song. For example, if a child was to sing a song in major tonality and the keyality
of D, he or she heard the d1 and a1 resonator bells. Next, the researcher turned
on the digital voice recorder and sang the appropriate major preparatory
sequence using a neutral syllable and gestured for the child to begin singing.
The digital voice recordings were stored on a computer in one of three ﬁles
(major, minor, improvisations). The recordings in each of the three ﬁles were
randomly rearranged by “dragging” a child’s recording into a different position

and “dropping” it into place. In so doing, not only was the presentation order

102

different from the original, it was also different for each of the three files. The
rearranged ﬁles were then copied onto Maxell CD-R compact discs.

To control for the effect of presentation order, some of the children
improvised ﬁrst, followed by the singing of the rote songs. Some children sang
the rote songs ﬁrst, followed by the improvisation. Other children sang one rote
song, followed by the improvisation, and then another rote song. Furthermore,
the presentation order of the rote songs alternated between the two tonalities.
The digital voice recordings were divided so that there was one digital voice ﬁle
containing the major tonality rote song performances, a separate ﬁle containing
the minor tonality rote song performances, and a third set of ﬁles containing the
Children’s improvisations. For each of the three separate digital voice ﬁles, the
treatment and control groups were intermingled and the presentation order was
rearranged for each ﬁle in order to eliminate as much bias as possible. Compact
disc recordings were produced from each of the three digital voice ﬁles.

Three independent judges reviewed the compact disc recordings and
rated the children’s song performances'using the Tonal Rating Scale and the
children’s improvisations using the Improvisation Rating Scale. The judges
received training in the use of the two criterion measures beforehand. To
increase judge objectivity in the rating of the student's performances, the judges
had no knowledge of the background and speciﬁc ages of the students. Each
judge received three compact discs, one containing the major rote song
performances, one containing the minor rote song performances, and another

containing the improvisation performances.

103

Research Instruments

Three criterion measures were used for this study. The Tonal subtest of
Intermediate Measures of Music Audiation (IMMA) (Gordon, 1982) served as a
pretest and posttest measure of the developmental tonal aptitude of all students
participating in this study.

The Primary Measures of Music Audiation (Gordon, 1979) is typically
given to kindergarten and ﬁrst-grade students. However, after several years of
administering PMMA to students at this particular school, the researcher found
that the majority of scores fall in the 90th to 99th percentile range. Therefore, she
elected to use IMMA instead. IMMA is designed for those students who score
exceptionally high on PMMA (Gordon, 1986).

The Tonal subtest of Intermediate Measures of Music Audiation (IMMA)
consists of 40 pairs of tonal patterns each containing three pitches. The patterns
are performed with a consistent tempo and without rhythm. All patterns are in
the keyality of C and in major or harmonic minor tonalities.

The students are asked to listen to the test recording and then decide if
each of the paired patterns (or songs as they are called in the test) are the same
or different. The paired patterns are identiﬁed on the recording by the name of
an object — “pencil” or “book,” for example. Next the students hear the word
“ﬁrst,” followed by the ﬁrst pattern of the pair and then the word ”second,”
followed by the second pattern of the pair. The object name given to the paired
patterns corresponds to a picture above a set of boxes on the test answer sheet

(see Appendix K). The top box of the set has two faces that look the same (two

104

smiling faces), and the bottom box has two faces that look different (one smiling
face and one frowning face). The students mark the answer Sheet by drawing a
circle around the box on top with the two faces that look the same if they think
that the two parts of the song sounded the same. If the students think the two
parts of the song were different, they are to draw a circle around the bottom box
with the two faces that are different.

A second measure was used to rate the children’s performance of two
criterion songs. The Tonal Rating Scale (see Appendix L) was developed based
on Levinowitz’s (1987) Tonal Rating Scale. Levinowitz’s scale is a ﬁve-point
continuous rating scale used to assess children’s singing achievement. The
scale consists of two dimensions — tonal and rhythm. When used by Levinowitz,
the interjudge reliabilities for ratings of kindergarten children’s major and minor
songs combined was .89. lnterjudge reliabilities for ﬁrst grade children (r = .94)
were slightly higher. The Tonal Rating Scale used for this study is similar to the
Levinowitz scale in that it is a ﬁve-point continuous rating scale with two
dimensions. The tonal dimensions of the two scales have similar criteria. The
difference lies in the second dimension. Because rhythm was not a concern of
this study, the second dimension focused on the children’s ability to remain in a
given keyality.

A third measure was used to rate the children’s tonal improvisations. The
Improvisation Rating Scale (see Appendix M) was developed by the researcher

and used to measure the tonal strength of the children’s improvisations.

105

Data Analysis

Means and standard deviations were computed for pretest and posttest
IMMA Tonal subtest scores and for scores on the Tonal Achievement Rating
Scale and the Improvisation Rating Scale. Means and standard deviations were
computed for each treatment, each level, and for the separate age levels within
the treatment and control groups. In addition, Split-halves reliabilities corrected
for length with the Spearrnan-Brown Prophecy formula were calculated for IMMA
Tonal subtest scores of the pretest and posttest administration. The interjudge
reliabilities of the Tonal Achievement and Improvisation Rating Scales were
calculated as well.

In order to establish whether there were preexisting differences between
mean scores split by grade on IMMA Tonal subtest, an unpaired ttest was
conducted. To determine the affect of root melody accompaniment on the
developmental tonal aptitude of kindergarten and ﬁrst-grade children, a two-way
analysis of variance (treatment x grade level) was conducted to determine
whether any significant differences exist between the two treatment groups or
levels. If signiﬁcant differences occurred favoring the use of a harmonic context
when teaching songs, ﬁndings would add to the body of research that instruction
inﬂuences children's developmental tonal aptitude. There may be greater gains
in aptitude when root melody accompaniment is used with song instruction. If no
signiﬁcance is found, several speculations would be made on the possibility of

song instruction including both songs with and without root melody

106

accompaniment. It may be possible that aptitude is unaffected by the
simultaneous presentation of melody and harmony.

To determine the affect of root melody accompaniment on the vocal
achievement of kindergarten and ﬁrst-grade children, a two-way analysis of
variance (treatment x grade level) was conducted to determine whether any
signiﬁcant differences exist between the two treatment groups or levels. If
signiﬁcant differences occurred favoring the use of a harmonic context when
teaching songs, ﬁndings would add to the body of research that children sing
more accurately when root melody accompaniment is used during instruction. If
no signiﬁcance is found, several speculations would be made on the possibility of
song instruction including both songs with and without root melody
accompaniment. It may be possible that children cannot focus on more than one
thing at a time, and the Simultaneous presentation of melody and harmony is
more information than young Children can digest at one time.

To determine the affect of root melody accompaniment on the vocal
improvisation of kindergarten and first-grade children, a two-way analysis of
variance (treatment x grade level) was conducted to determine whether any
signiﬁcant differences exist between the two treatment groups or levels. If
signiﬁcant differences occurred favoring the use of a harmonic context when
teaching songs, findings would add new information to the research literature
suggesting that children’s vocal improvisations are more cohesive when root
melody accompaniment is used during instruction. That is, children’s

improvisations convey a greater sense of being constructed around a key center

107

and a more reﬁned sense of ﬁnality or conclusion. If no signiﬁcance is found,
several speculations would be made on the possibility of song instruction
including both songs with and without harmonic accompaniment. Alpha was set

at .05 level for all analyses.

108

CHAPTER FOUR

RESULTS AND INTERPRETATIONS

Reliabilities of Pre-lnstruction and Post-Instruction Tonal sgbtest of Intermediate
Measures of Music Audiation

Split-half reliabilities were computed for the present study using the pre-
instruction and post-instruction administration of the Tonal subtest of IMMA for
the kindergarten and ﬁrst grade Children. The split-halves reliability coefﬁcients
for the pretest and posttest of IMMA for ﬁrst-grade in the present study were
comparable to the ones reported in the IMMA Manual (Gordon, 1986). The split-
halves reliability coefﬁcient reported in the IMMA Manual for ﬁrst-grade students
for the Tonal subtest is .76. The split halves reliabilities, corrected using
Spearman Brown Prophecy, for the pre-instruction administration were .76 for
children in kindergarten and .83 for children in the ﬁrst grade. The split halves
reliabilities, corrected using Spearman Brown Prophecy, for the post-instruction
administration were .83 for children in kindergarten and .66 for children in the first

grade.
Reliabilities of Tonal Achievement Rating Scale

At the conclusion of the 25-week instructional period, three independent

judges rated the students’ performances of the two criterion songs (see Tables 2

109

and 3). The interjudge reliabilities were as follows: Tonality Dimension for the
major criterion song ranged from r = .76 to .84; Tonality Dimension for the minor
criterion song ranged from r = .77 to r = .79; Keyality Dimension for the major
criterion song ranged from r = .69 to r = .79; Keyality Dimension for the minor
criterion song ranged from r = .66 to r = .80; major tonality composite reliabilities
ranged from r = .83 to r = .87; minor tonality composite reliabilities ranged from
r = .83 to r = .89, and; composite tonal achievement reliabilities ranged from r =
.88 to r = .93. The reliability coefficients ranged from moderate to high, and the

ratings appear to be reliable.

110

Table 2

lnterjudge Reliabilities for Rating Scales

 

Maior - Tonality Dimension

 

 

 

 

 

 

Judge 1 Judge 2 Judge—3

Judge 1 1.00 .77 .76
Judge 2 .77 1.00 .84
Minor - Tonality Dimension

Judge 1 1.00 .79 .78
Judge 2 .79 1.00 .77
Maior - Keyality Dimension

Judge 1 1.00 .79 .73
Judge 2 .79 1.00 .69
Minor - Kevalitv Dimension

Judge 1 1.00 .70 .80
Judge 2 .70 1.00 .66
Maior — Composite Score

Judge 1 1.00 .87 .85
Judge 2 .87 1.00 .83
Minor — Composite Score

Judge 1 1.00 .85 .89
Judge 2 .85 1.00 .83

111

Table 3

lnterjudge Reliabilities for Composite Tonal Achievement Score

 

 

Judge 1 Judge 2 Judge 3
Judge 1 1.00 .91 .93
Judge 2 .91 1.00 .88

 

Reliabilities of the Improvisation Rating Scale
At the conclusion of the 25-week instructional period, three independent
judges rated the students’ improvisations (see Table 4). The interjudge reliability

ranged from .74 to .80 for the improvisation rating scale.

Table 4

lnterjudge Reliabilities for me Improvisation Rating Scale

 

Judge 1 Judge 2 Judge 3
Judge 1 1.00 .74 .76

Judge 2 .74 1.00 .80

 

112

The Effect of Root Melody Accompaniment to Soggs on the Developmental
Tonal Aptitude of Children in Kindergarten and First Grade

Unpaired tTest of IMMA Pre-lnstruction Tonal Scores

In order to establish whether there were preexisting differences between
mean scores split by grade on IMMA Tonal subtest, means and standard
deviations were calculated (see Table 5) as well as an unpaired ttest (alpha of
.05). There was no signiﬁcant difference between the experimental and control
groups with grade levels combined, between the experimental and control groups
for Children in kindergarten, or between the experimental and control groups for

children in first grade (see Table 6).

Table 5

Groupings of Pre-Instwtion Tonal Scores Between Treatment Grogps and Split

 

 

by Grade

N M SD
Experimental: Total 68 30.71 4.66
Control: Total 68 30.40 4.44
Experimental: SK 34 29.65 4.77
Control: SK 34 28.38 4.24
Experimental: Ist 34 31.77 4.36

Control: 1st 34 32.41 3.69

 

113

Table 6
Unpaired tTest of Pre-lnstruction Tonal Scores Between Treatment Groups and

SJIIt by Grade

 

 

MD df t D
Experimental, Control: Total -.31 134 -.40 .69
Experimental, Control: SK -1.65 66 -1.16 .25
Experimental, Control: 1st -.65 66 -.66 .51

 

p>.05

Mgans anti Standard [Leviations of Tonal subtest of IMMA Post-lnstrgction
§29_re§

Means and standard deviations for the post-instruction scores of the Tonal
subtest of IMMA are presented in Table 7. The experimental group with grade
levels combined received slightly higher tonal aptitude scores than the control
group with grade levels combined. The scores of the kindergarten experimental
group were slightly higher than those of the kindergarten control group on the
post-instruction administration of IMMA. The scores of the ﬁrst-grade
experimental group and the ﬁrst-grade control group were similar.

The standard deviation for the experimental group with grade levels
combined was similar to that of the control group with grade levels combined,
indicating similar variability of scores between the two groups. The standard

deviation for the kindergarten experimental group was comparable to that of the

114

kindergarten control group, indicating similar variability of scores between the two
groups. The standard deviation for the ﬁrst-grade experimental group was
comparable to that of the ﬁrst-grade control group, indicating similar variability of

scores between the two groups.

Table 7

Means Table of Post-Instruction Developmental Tonal Aptitude Scores

 

 

N M SD
Experimental Total 68 33.66 3.85
Control Total 68 32.77 3.79
SK, Experimental 34 33.06 4.40
SK, Control 34 31.44 4.12
1st, Experimental 34 34.27 3.16
1st, Control 34 34.09 2.93

 

ANOVA of Post-Instruction IMMA Tonal Scores

In order to determine whether the use of a root melody accompaniment
with song instruction has an effect on developmental tonal aptitude, a two-way
analysis of variance (treatment x grade level) was conducted (p 5 .05). Results
(see Table 8) indicated no significant interactions or differences as a result of

treatment. However, there was a signiﬁcant difference between levels. As would

115

be expected, children in the ﬁrst grade had signiﬁcantly higher tonal aptitude

scores than children in kindergarten.

Table 8

ANOVA Table of Posttest Developmental Tonal Aptitude Scores

 

 

df SS MS F p
Grade 1 126.18 126.18 9.20 .00"
Treatment 1 27.36 27.36 1.99 .16
Treatment x Grade 1 17.65 17.65 1.29 .26
Residual 132 1809.62 13.71

 

*p < .05

Interpretation of IMMA Tonal Aptitude Scores

In this study treatment had no effect on the developmental tonal aptitudes
of young children. However, children in ﬁrst grade did better than children in
kindergarten. Children’s developmental aptitude scores increase as a result of
age, regardless of treatment.

An explanation for the lack of differences due to treatment may be that the
Tonal subtest of IMMA was too easy for the children in this study. The post-
instruction tonal aptitude mean scores are well above the mean reported in the
test manual (Gordon, 1986). Gordon reports a mean of 31.1 for children in ﬁrst
grade. The aptitude mean scores of the children in this study ranged from 31.44

(kindergarten control group) to 34.27 (ﬁrst-grade experimental group). This may

116

have resulted in the lower split-halves reliability (.66) of the post-instruction
administration for children in ﬁrst grade. When too many children receive
relatively the same score, reliability decreases. It was apparent going into this
study that IMMA might not be ideal for this particular group of Children. However,
because no more difﬁcult and age-appropriate developmental aptitude test
exists, there was no other choice.

Another possibility is that children in this study received two years of
informal music guidance prior to formal music instruction. The environment has
an affect on children's aptitude until about age nine (Gordon, 1980b, 1989,
19970, 19983, 1999; Taggart, 1997). The high aptitude levels found in this study
may be, in part, the result of the rich and varied listening vocabulary children
gained as a result of this informal guidance. Another aspect of informal guidance
and formal music instruction that may have supported the developmental tonal
aptitude of the children in the present study is individualized instruction. Knowing
the tonal strengths and weakness of individual Children, and individualizing
instruction to meet these needs, helps children to realize their music potential.
Because these children's aptitudes were so high to begin with, it is possible that
there was not sufﬁcient room for growth as a result of treatment.

Parental involvement in music is extremely high at this particular school.
Gordon (1976) investigated the relationship of environmental factors to both
music aptitude and achievement test scores. He found that that several factors
(e.g., ”siblings sing or play" and "a piano at home") had great inﬂuence on a

child's achievement. Many parents report that they sing and play music for their

117

children on a regular basis. Several of the parents are professional musicians in
major orchestras, and a few are well-known composers. In addition, the
importance of the arts is recognized in the school's mission statement.
Furthermore, the majority of students who attend this school come from an
upper-middle or high socio-economic class. Participation in music activities is
encouraged and supported through the home and school environment, and, thus,
music plays a major role in the life of children at this school. This is not the case
for most schools across the country.

It is possible that 25 weeks of song instruction using root melody
accompaniment was not enough to affect children’s developmental tonal
aptitude. Although the differences between the combined experimental group
and combined control group were not signiﬁcant, the mean scores tended to be
higher for the combined experimental group. Furthermore, the kindergarten
experimental group tended to receive higher mean scores than the kindergarten
control group. The same trend also occurred between the ﬁrst-grade
experimental group and the ﬁrst-grade control group. A longer treatment period
may have yielded different results.

In the analysis of the pre-instruction and post-instruction IMMA tonal
aptitude scores by age-level, it was found that children in kindergarten scored
signiﬁcantly lower than the Children in ﬁrst grade. Because older children score
better than younger children on IMMA as a result of maturation (Gordon, 1986),
the overall difference between the two groups in pre- and post-instruction IMMA

mean scores are attributable to the age difference between the two groups. It is

118

also possible that the difference in the amount of instruction between the two
grade levels may have contributed to this finding. Children in kindergarten
received one hour less of instruction per cycle and one full year less of formal

music instruction than Children in ﬁrst grade.

TheEffect of Root Melody Accompaniment to Sopgs on the Tonal Achievement
of Children in Kindergarten and First Grade

Means and Standard Deviations of Tonal Ratigg Scale — Tonality Dimension

The means and standard deviations of Tonality Dimension scores for the
major and minor criterion songs are reported in Table 9. The combined
experimental group major criterion song mean and minor criterion song mean are
similar to, but slightly higher than, the combined control group mean. Similar
means between grade levels and within grade levels were also found, with the
experimental tendencies slightly higher than those of the control group except for
the major song in grade one. Children in kindergarten received slightly lower
scores than children in the ﬁrst grade on both the major and minor criterion
songs. The mean tonality score of the kindergarten experimental group was
slightly higher than that of the kindergarten control group for the major and minor
criterion songs. The mean trends of the ﬁrst-grade experimental group and the
ﬁrst-grade control group were mixed.

The standard deviation for the experimental group with combined grade

levels was similar to that of the control group for combined grade levels,

119

indicating similar variability of scores between the two groups for the major,
minor, and composite tonality scores. The standard deviation for the
kindergarten experimental group was comparable to that of the kindergarten
control group, indicating similar variability of scores between the two groups.
The standard deviation for the first-grade experimental group was slightly larger
than that of the ﬁrst-grade control group, indicating a greater variability of scores

in the experimental group.

120

Table 9

Means Table of Tonality Scores

 

 

 

 

N M SD
Major Tonality
Experimental Total 68 3.66 1.05
Control Total 68 3.56 1.02
SK, Experimental 34 3.64 .95
SK, Control 34 3.23 .99
1st, Experimental 34 3.68 1 .14
Ist, Control 34 3.88 .94
' Minor Tonality
Experimental Total 68 3.61 1.03
Control Total 68 3.46 1.02
SK, Experimental 34 3.47 1.02
SK, Control 34 3.25 1.06
lst, Experimental 34 3.76 1.04
1 st, Control 34 3.67 .95

 

The means and standard deviations of the tonality composite scores are
reported in Table 10. The combined experimental group tonality composite
mean is slightly higher than the control group mean. The tonality composite

mean of the kindergarten experimental group was slightly higher than that of the

121

kindergarten control group. The mean of the ﬁrst-grade experimental group was

slightly lower than that of the ﬁrst-grade control group.

Table 10

Means Table of Tonality Composite Score

 

 

N M SD
Experimental Total 68 7.28 1.94
Control Total 68 7.02 1.91
SK, Experimental 34 7.1 1 1.88
SK, Control 34 6.48 1.93
Ist, Experimental 34 7.44 2.02
1 st, Control 34 7.56 1 .77

 

The standard deviation for the experimental group with combined grade
levels was similar to that of the control group, with combined grade levels
indicating similar variability of scores between the two groups. The standard
deviation for the kindergarten experimental group was comparable to that of the
kindergarten control group, indicating similar variability of scores between the two
groups. The standard deviation for the ﬁrst-grade experimental group was
somewhat larger than that of the ﬁrst-grade control group, indicating a greater

variability of scores in the experimental group.

122

ANOVA of Tonality Scores

In order to determine whether the use of a root melody accompaniment in
song instruction has an effect on kindergarten and ﬁrst-grade children’s abilities
to sing tonic, dominant, subdominant and cadential patterns accurately within the
context of a song, a two-way analysis of variance (treatment x grade) was
conducted (p 5 .05). Results for the major and minor tonality mean scores (see

Table 11) and the tonality composite mean scores (see Table 12) indicated no

 

signiﬁcant interactions or differences as a result of treatment. As expected, there
was a signiﬁcant difference for the main effect of grade for the major tonality

scores, minor tonality scores, and tonality composite scores.

123

Table 11

_A_NOVA Table of Tonal RatingScale — Tonality Dimension

 

 

 

 

 

 

 

 

 

df 88 MS F D

Major Tonality
Treatment 1 .37 .37 .36 .55
Grade 1 4.06 4.06 3.97 .05*
Treatment x Grade 1 3.15 3.15 3.08 .08
Residual 132 135.02 1.02
Minor Tonality
Treatment 1 .78 .78 .75 .39
Grade 1 4.34 4.34 4.19 .04*
Treatment x Grade 1 .15 .15 .14 .71
Residual 132 136.77 1.04
*p 5 .05
Table 12
ANOVA Table of Tonality Composite Score

df SS MS F 0
Treatment 1 2.23 2.23 .62 .43
Grade 1 16.80 16.80 4.65 .03*
Treatment x Grade 1 4.67 4.67 1.29 .26
Residual 132 476.72 3.61
*p _<_ .05

124

Interpretation of the Tonality Score

Results indicated that no significant difference existed between the
experimental and control groups. In this study, the addition of root melody
accompaniment to songs did not have an effect on children’s ability to sing tonic,
dominant, subdominant and cadential patterns accurately within the context of a
song. This finding supports the results of Atterbury 8 Silcox (1993), Petzold
(1966), and Stauffer (1986) that harmonic accompaniment and instruction does
not have an affect on children's singing accuracy.

The children in this study have had two years of informal music guidance
that included rich listening experiences in all tonalities. It is possible that the
acculturation process is sufﬁcient in developing children's ability to sing within a
tonality, so that the treatment had no effect.

Longer instructional periods allow children to achieve closer to their
potential. Perhaps 25 weeks of root melody accompaniment with song
instruction was not enough to affect children’s ability to sing within a tonality, so
again, the treatment had no effect. A longer treatment period may have resulted
in signiﬁcant differences.

Another possible reason that the addition of root melody accompaniment
to song instruction did not have an effect on children’s ability to accurately Sing
within a tonality of a song may have to do with developmental maturity.
Children’s ability to sing accurately within the tonality of a song may still be in the

process of developing. If this is the case, the affects of root melody

125

accompaniment with song instruction will not be obvious until later in the
developmental process.

From the analysis of the tonality scores, it was found that children in
kindergarten scored signiﬁcantly lower than the children in first grade for the
combined control and experimental group and the control group alone. A factor
that could be contributing to the difference is that older children score better than
younger children by the advantage of age. In addition, children in ﬁrst grade had
one full year more of instruction than children in kindergarten. Furthermore, the
difference in the amount of instruction received by the two grade levels may have
contributed to this ﬁnding. Kindergarten children received one hour less of

instruction per cycle than children in ﬁrst grade.

Means and Standard Deviations of Tonal Rating Scale — Keyality Dimension

The means and standard deviations of the Keyality Dimension scores for
the major and minor criterion songs are reported in Table 13. The combined
experimental group major criterion song mean is similar to the combined control
group mean scores. The combined experimental group minor criterion song
mean is slightly higher than the combined control group mean scores. Similar
means between grade levels and within grade levels were also found. Children
in kindergarten received relatively the same score as children in the ﬁrst grade on
both songs. The keyality scores of the kindergarten experimental group were

comparable to the kindergarten control group for the major song and were

126

slightly higher for the minor song. The means of the ﬁrst-grade experimental

group and the ﬁrst-grade control group were similar.

Table 13

Means Table of Tonal Rating Scale — Kevalitngimension

 

 

 

 

N M SD

Maior Kevalitv

Experimental Total 68 2.21 1.36
Control Total 68 2.19 1 .42
SK, Experimental 34 2.12 1.22
SK, Control 34 2.06 1.37
1st, Experimental 34 2.29 1.49
1st, Control 34 2.33 1 .47
Minor Kevalitv

Experimental Total 68 2.29 1.28
Control Total 68 1.92 1 .19
SK, Experimental 34 2.35 1.31
SK, Control 34 1.73 1.12
1 st, Experimental 34 2.22 1 .27
1 st, Control 34 2.12 1 .24

 

127

The standard deviation for the experimental group with combined grade
levels was similar to that of the control group with combined grade levels,
indicating similar variability of scores between the two groups. The standard
deviation for the kindergarten and ﬁrst-grade experimental groups was
comparable to that of the kindergarten and ﬁrst-grade control groups, indicating
similar variability of scores between the groups.

The means and standard deviations of the Keyality Composite scores are
reported in Table 14. The combined experimental group composite keyality
means are similar to the combined control group means, with a slight trend

favoring the experimental group.

Table 14

Means Table of Kevalitv Composite Score

 

 

N M SD
Experimental Total 68 4.49 2.43
Control Total 68 4.19 2.28
SK, Experimental 34 4.47 2.37
SK, Control 34 3.79 2.30
1 st, Experimental 34 4.52 2.52
1 st, Control 34 4.45 2.26

 

128

Similar means between grade levels and within grade levels were also found,
again with slight trends favoring the experimental and control groups and the
older children. Children in kindergarten received nearly the same keyality
composite scores as children in the ﬁrst grade. The keyality composite mean of
the kindergarten experimental group was slightly higher than that of the
kindergarten control group. The mean of the ﬁrst-grade experimental group and
the ﬁrst-grade control group were also similar. All mean trends, although slight,
favored the experimental group.

The standard deviation for the experimental group with combined grade
levels was similar to that of the control group with combined grade levels,
indicating similar variability of scores between the two groups. The standard
deviations for the kindergarten and ﬁrst-grade experimental groups were
comparable to the ones for the kindergarten and ﬁrst-grade control groups,

indicating similar variability of scores between the groups.

ANOVA of Kevalitv Scores

In order to determine whether the use of a root melody accompaniment in
song instruction has an effect on kindergarten and ﬁrst-grade children’s ability to
remain in the given keyality of a song, a two-way analysis of variance (treatment
x grade) was conducted, and alpha was set at .05. Results for the major and
minor criterion song ratings (see Table 15) and the keyality composite ratings
(see Table 16) yielded no signiﬁcant interactions or differences for the main
effects of treatment or grade level. However, the difference between groups as a

result of treatment approached signiﬁcance for the minor keyality song.

129

Table 15

ANOVA Table of Tonal Rating Scale — Kevalitv Dimension

 

 

 

 

 

 

 

 

 

df 88 MS F p

Maior Kevalitv
Treatment 1 .01 .01 .00 .95
Grade 1 1.61 1.61 .83 .36
Treatment x Grade 1 .08 .08 .04 .84
Residual 132 256.53 1.94
Minor Kevalitv
Treatment 1 4.45 4.45 2.91 .09
Grade 1 .60 .60 .39 .53
Treatment x Grade 1 2.33 2.33 1.52 .22
Residual 132 201.76 1.53
p > .05
Table 16
ANOVA Table of Kevalitv Composite Score

df S_S MS F p
Treatment 1 4.82 4.81 .86 .35
Grade 1 4.17 4.17 .75 .39
Treatment x Grade 1 3.24 3.24 .58 .45
Residual 132 737.51 5.59
p > .05

130

Interpretation of the Kevalitv Scores

No signiﬁcant difference existed between the experimental and control
groups or between grade levels. The addition of a root melody accompaniment
in song instruction did not have an effect on children’s ability to remain in the
given keyality of a song. This ﬁnding supports the results of Atterbury & Silcox
(1993), Petzold (1966), and Stauffer (1986), who found that harmonic
accompaniment does not have an affect on children's singing accuracy.

As with the previous interpretations, the children in this study have had
two years of informal music guidance that included rich listening experiences in
all tonalities. It is possible that the acculturation process is sufficient in
developing children's ability to sing within the keyality of a song, regardless of
treatment.

The minor keyality mean difference approached signiﬁcance as a result of
treatment. Children in the experimental group tended to receive higher ratings
than children in the control group. Furthermore, the kindergarten experimental
group tended to receive higher scores than the kindergarten control group. The
same trend also occurred between the ﬁrst-grade experimental group and the
ﬁrst-grade control group. A longer study investigating the effect of root melody
accompaniment on children’s ability to remain in the given keyality of a song
might have yielded signiﬁcant results.

From the analysis of the keyality scores, it was found that children in
kindergarten received the same keyality mean scores as children in ﬁrst grade -

surprisingly, age appears not to be a factor. A possible explanation may be that

131

it is difﬁcult for young children to remain in the given keyality of a song. The
children's major and minor keyality means were lower than their tonality mean.
They could audiate tonality but may need more experience before developing the
ability to audiate keyality. Many of the children, when performing the criterion
songs, were able to begin the songs in the given keyality but then strayed from it
during the B section. Some children were able to regain the keyality again during
the ﬁnal A section of the songs. This conﬂicts with Gordon (1997a, 1997c), who
states that children develop the ability to sing within the keyality of songs before
they accurately sing within the tonality of songs.

Another explanation for the similarity of keyality scores between grade
levels may be that the keyality dimension of the rating scale needs reﬁnement, as
it had the lowest reliabilities of all of the rating scales. This dimension of the
rating scale could have been difﬁcult for the judges and therefore did not serve
as an adequate measure of the children's ability.

Another possibility for the similarity of keyality scores between grade
levels is that there were different instructors for one of the grade levels. One
teacher provided instruction for the children in all of the kindergarten classes and
two of the ﬁrst-grade classes. Two other teachers provided instruction for the
children in one of the remaining two ﬁrst-grade classes. Because no two
teachers can teach exactly alike, the difference in instruction, as it related to
keyality, may have been a factor. Therefore, no keyality performance difference

between grade levels was found.

132

Means and Standard Deviations for Major and Minor Criterion m Composite
m

The means and standard deviations of the major and minor criterion song
composite scores are reported in Table 17. The combined experimental group
major and minor criterion song composite scores are slightly higher than those of
the control group. Similar means between grade levels were also found, with a
slight trend in favor of the older children. Children in kindergarten received
nearly the same major and minor song composite scores as children in the ﬁrst
grade. The major and minor song means of the kindergarten experimental group
were slightly higher than those of the kindergarten control group. The means of
the ﬁrst-grade experimental group and the ﬁrst-grade control group were similar,

slightly favoring the control group.

133

Table 17

Means Table for Maior and Minor Criterion Song Composite Scores

 

 

 

 

N M 812
MaLor SongComposite
Experimental Total 68 5.89 2.21
Control Total 68 5.75 2.27
SK, Experimental 34 5.77 2.00
SK, Control 34 5.29 2.26
1st, Experimental 34 5.99 2.43
1st, Control 34 6.21 2.22
Minor Song Composite
Experimental Total 68 5.91 2.21
Control Total 68 5.39 2.05
SK, Experimental 34 5.84 2.26
SK, Control 34 4.95 2.02
1st, Experimental 34 5.98 2.19
1st, Control 34 5.82 2.01

 

The standard deviation for the experimental group with combined grade
levels was similar to that of the control group with combined grade levels,
indicating similar variability of scores between the two groups. The standard

deviations for the kindergarten and first-grade experimental groups were

134

comparable to the ones for the kindergarten and ﬁrst-grade control groups,

indicating similar variability of scores between the groups.

ANOVA of Maior and Minor Criterion Song Composite Scores

In order to determine whether the use of a root melody accompaniment in
song instruction has an effect on kindergarten and ﬁrst-grade children's ability to
sing a major and minor criterion song, a two-way analysis of variance (treatment
x grade) was conducted, and alpha was set at .05. Results for the major and
minor criterion song composite scores (see Table 18) indicated no signiﬁcant

interactions or difference for the main effects of treatment or grade level.

135

Table 18

ANOVA Table of Major and Minor Criterion Song Composite Scores

 

 

 

 

df SS MS F p

Maior Song Composite

Treatment 1 .45 .45 .09 .77
Grade 1 10.62 10.62 2.13 .15
Treatment x Grade 1 4.31 4.31 .86 .35
Residual 132 658.40 4.99

Minor SongComposite

Treatment 1 9.42 9.42 2.09 .15
Grade 1 8.70 8.70 1.93 .17
Treatment x Grade 1 4.67 4.67 1.04 .31
Residual 132 595.63 4.51

 

p>.05

Interpretation of Maior and Minor Criterion Songgomposite Scores

No signiﬁcant difference existed between the experimental and control
groups or between grade levels. In this study, the addition of a root melody
accompaniment in song instruction did not have an effect on children’s ability to
sing a major and minor song accurately. This ﬁnding supports the results of
Atterbury & Silcox (1993), Petzold (1966), and Stauffer (1986), who found that

harmonic accompaniment does not have an affect on children's singing accuracy.

136

Again, the children in this study have had two years of informal music
guidance that included rich listening experiences in all tonalities. It is possible
that the acculturation process is sufﬁcient in developing children's ability to sing
within the tonality of a song, regardless of treatment.

From the analysis of the major and minor criterion song composite scores,
it was found that children in kindergarten received the same mean scores as

children in ﬁrst grade —- again, age appears not to be a factor.

Means and Standard Deviations for Composite Tonal Achievement Scores

The means and standard deviations of the composite tonal achievement
scores are reported in Table 19. The combined experimental group composite
tonal achievement scores are similar to the combined control group scores, with
a slight difference in favor of the experimental group. Similar means between
grade levels and within grade levels were also found. Children in kindergarten
received nearly the same composite tonal achievement score as children in the
ﬁrst grade, with a slight trend in favor of the older children. The composite tonal
achievement mean of the kindergarten experimental group was similar to that of
the kindergarten control group, with a trend in favor of the control group. The
means of the ﬁrst-grade experimental group and the ﬁrst-grade control group

were also similar.

137

Table 19

Means Table for Composite Tonal Achievement

 

 

N M SD
Experimental Total 68 1 1.78 4.16
Control Total 68 1 1.14 4.01
SK, Experimental 34 1 1.61 4.06
SK, Control 34 10.24 4.08
1st, Experimental 34 1 1.95 4.31
1st, Control 34 12.03 3.78

 

The standard deviation for the experimental group with combined grade
levels was similar to that of the control group with combined grade levels,
indicating similar variability of scores between the two groups. Standard
deviations for the kindergarten and ﬁrst-grade experimental groups were
comparable to the ones for the kindergarten and ﬁrst-grade control groups,

indicating similar variability of scores between the groups.

ANOVA of Composite Tonal Achievement Score

In order to determine whether the use of a root melody accompaniment in
song instruction has an effect on kindergarten and ﬁrst-grade children’s overall
tonal accuracy, a two-way analysis of variance (treatment x grade) was

conducted, and alpha was set at .05. Results for the composite tonal

138

achievement scores (see Table 20) indicated no signiﬁcant interactions or

difference for the main effects of treatment or grade level.

Table 20

ANOVA Table of Tonal Achievement

 

 

df SS MS F j
Treatment 1 13.98 13.98 .85 .35
Grade 1 38.54 38.54 2.34 .13
Treatment x Grade 1 17.94 17.94 1.09 .30
Residual 32 2177.63 16.50

 

p >.05

Interpretation of Tonal Achievement Composite Scores

No signiﬁcant difference existed between the experimental and control

groups or between grade levels. In this study, the addition of a root melody

accompaniment in song instruction did not have an effect on children’s tonal

accuracy. Although Jarjisian (1983), and Richner (1976) found signiﬁcant

difference in tonal achievement related to instruction, Levinowitz (1987) did not.

Furthermore, Atterbury & Silcox (1993), Petzold (1966), and Stauffer (1986)

found no difference in the tonal accuracy of children who sang songs within a

harmonic context and those who did not. It is possible that children cannot focus

on more than one aspect of music at a time when singing. The focus may have

139

been placed on the teacher's voice with little or no focus on the root melody
accompaniment.

It is also possible that when informal early childhood music experiences
provide a foundation of listening, children are able to sing more accurately later
on. The process of music learning parallels the process of language learning
(Gordon, 1990) - the larger the "listening" vocabulary, the larger the "speaking"
vocabulary. Stamou (1999) found that Suzuki students who had informal music
guidance in early childhood tended to receive higher ratings on intonation,
rhythm, expression, and composite performance achievement scores than
children who did not have such instruction. All children had completed two years
of music instruction prior to participation in this study.

From the analysis of the tonal achievement composite means, it was
found that children in kindergarten received the same scores as children in ﬁrst
grade - age again appears not to be a factor. Another possibility for the similarity
of tonal achievement scores between grade levels is that there were different
instructors for one of the grade levels. One teacher provided instruction for the
children in all of the kindergarten classes and two of the ﬁrst-grade classes. Two
other teachers provided instruction for the children in one of the remaining two
ﬁrst-grade classes. Because no two teachers can teach exactly alike, the
difference in instruction, as it related to keyality, may have been a factor.

Therefore, no keyality performance difference between grade levels was found.

140

The Effect of Root Melodv Accomﬁniment to Scms on the
Tonal Improvisations of Children in Kindergarten and First Grade

Means and Standard Deviations for Improvisation Scores

The means and standard deviations of the improvisation scores are
reported in Table 21. The combined experimental group improvisation scores
are higher than the combined control group scores. Also, children in
kindergarten received lower improvisation scores than children in ﬁrst grade. In
fact, the improvisation mean score of the kindergarten experimental group was
higher than that of the kindergarten control group. The mean score of the ﬁrst-
grade experimental group was higher than that of the ﬁrst-grade control group.

The standard deviation for the experimental group with combined grade
levels was similar to that of the control group with combined grade levels,
indicating similar variability of scores between the two groups. Standard
deviations for the kindergarten and ﬁrst-grade experimental groups were
comparable to those of the kindergarten and ﬁrst-grade control groups, indicating

similar variability of scores between the groups.

141

Table 21

Means Table for Improvisation Scores

 

 

N M SD
Experimental Total 68 2.61 1.43
Control Total 68 1 .97 1.41
SK, Experimental 34 2.61 1.43
SK, Control 34 1.97 1.41
1st, Experimental 34 3.11 1.43
1st, Control 34 2.58 1 .27

 

ANOVA of Tonal Improvisation Scores

In order to determine whether the use of a root melody accompaniment in
song instruction has an effect on kindergarten and ﬁrst-grade children’s tonal
improvisations, a two-way analysis of variance (treatment x grade) was
conducted, and alpha was set at .05. Results for the tonal improvisation scores
(see Table 22) indicated no signiﬁcant interactions. However, there were
statistically signiﬁcant differences for the main effects of treatment and grade
level. Children in the experimental group received signiﬁcantly higher
improvisation ratings than children in the control group. The mean scores of
children in first-grade were signiﬁcantly higher than those of children in

kindergarten.

142

Table 21

ANOVA Table of Improvisation Scores

 

 

df SS MS F 9
Treatment 1 11.71 11.71 6.08 .01*
Grade 1 10.45 10.45 5.43 .02*
Treatment x Grade 1 .12 .12 .06 .80
Residual 32 254.23 1.93

 

*p 5 .05

Interpretation of Improvisation Scores

Children who received song instruction with root melody accompaniment
achieved signiﬁcantly higher tonal ratings when improvising than children who did
not have such instruction. Implied harmonic functions within the original keyality
and tonality occurred most often in the melodic improvisations of children in the
experimental group. It is possible that, as it is with developing a listening
vocabulary of tonalities, children need exposure to the sound of harmonic
progressions within the context of familiar songs in order to build a harmonic
"speaking" vocabulary. An observation made by one of the teachers participating
in the study was that it was quite different hearing and singing the songs with the
root melody accompaniment. She felt that the accompaniment "brought the
song to life" in the sense that when she sang the song without the root melody,
she could clearly hear the progression within the melody. Teachers of young

children are aware of children's appearance that they are not paying attention,

143

but then those same children retain and display every bit of information at a later
time. This may be the case with improvisation. Although attention was never
brought to the root melody accompaniment, children were able to bring that
aspect into their improvisations. Listening to the bass line helped children
develop a sense of how melodies combine to create harmony and how tones
function within the harmony. It is familiarity and comfort with chord changes that
is the compelling factor in children's ability to improvise (Gordon, 2001).

From the analysis of the improvisation scores, it was found that children in
kindergarten scored signiﬁcantly lower than the children in ﬁrst grade. A factor
that could be contributing to the difference is that, because older children score
better than younger children by the advantage of age alone, the overall
difference between the two groups in improvisation mean scores are attributable
to maturation. Or, it may have been because children in kindergarten received
one hour less of instructional time per cycle than children in ﬁrst grade. Children
in kindergarten also had one full year less of formal music instruction than

children in ﬁrst grade.

144

CHAPTER FIVE

SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS

A summary of the study is presented in this chapter. Conclusions and
implications for music education are also included, followed by recommendations

for future research.

Summary of the Study
Eu_rgose and Problems
The purpose of this research was to examine the effect of harmonic
accompaniment on the developmental tonal aptitude, tonal achievement, and
tonal improvisations of young children. The speciﬁc problems of this study were
the following:
1) Does the addition of a root melody accompaniment to
song instruction affect the developmental tonal aptitude of
children in kindergarten and first grade?
2) Does the addition of a root melody accompaniment to
song instruction affect the tonal achievement of children in
kindergarten and ﬁrst grade?
3) Does the addition of a root melody accompaniment to
song instruction affect the tonal strength of the improvisations

of children in kindergarten and ﬁrst grade?

145

Design and Procedures

The design of this study was treatment by levels. The two treatments
were songs with no harmonic accompaniment and songs with harmonic
accompaniment. The two levels were grade in school (kindergarten and ﬁrst).

Sixty-eight kindergarten students and 68 ﬁrst grade students (N = 136)
from an independent private school participated in the study. The children in this
study were from four intact kindergarten classes (17 students per class) and four
intact ﬁrst grade classes (17 students per class). The students received a total of
twenty-ﬁve weeks of instruction. Children in kindergarten received 30-minutes of
formal music instruction twice per six-day instructional cycle. The children in ﬁrst
grade received 40-minutes of formal music instruction three times per cycle.

Music instruction for all children in this study was based on Gordon's
Music Learning Theory and included pattern instruction, identiﬁcation of tonalities
and meters, identiﬁcation of pattern functions, improvisation, singing, chanting,
movement, and playing instruments. For the treatment group (two kindergarten
and two ﬁrst grade classes), a root melody accompaniment was used with the
rote songs in each instructional period. The control group (two kindergarten and
two ﬁrst grade classes) did not have root melody accompaniment with the rote
songs sung in class. Other than the use of root melodies, all children received
the same instructional content.

The Intermediate Measures of Music Audiation was administered prior to

instruction and then again after instruction ended. Both the Tonal and Rhythm

146

subtests were administered but only the Tonal subtest was used as a dependent
variable.

The two criterion songs were taught to all children four weeks before
testing began. Within two weeks following the completion of instruction, each
individual student met with the researcher to perform the two criterion songs
without text, one in major tonality/duple meter and one in minor tonality/duple
meter, and to improvise an ending to a major tonality/duple meter song without
text. All singing was performed without root melody accompaniment. All of the
performances were audio-taped and used for analysis.

Three independent judges were used to rate the children's singing. They
rated both of the criterion songs using a ﬁve-point continuous rating scale with
two dimensions - tonality and keyality. The improvisations were rated using a
ﬁve-point continuous rating scale. All of the judge’s ratings were used in the

analysis.

Results and Interpretations

Split-halves reliabilities for the pre- and post-instruction administration of
the Tonal subtest of IMMA were, in general, slightly higher than the ones
reported in the IMMA Manual (Gordon, 1986). However, split-halves reliability for
the ﬁrst-grade post-instruction administration of IMMA was slightly lower than the
ones reported in the IMMA Manual. lnterjudge reliabilities between the three

judges ranged from .66 to .93 for the Tonal and Improvisation rating scales. The

147

reliability coefﬁcients ranged from moderate to high, and the ratings were
considered reliable.

In this study, song instruction with root melody accompaniment did not
have an effect on the developmental tonal aptitude of children in kindergarten
and ﬁrst-grade. No signiﬁcant difference was found between children in the
treatment group and the control group. This could be a result of several factors.
Perhaps the Tonal subtest of IMMA was too easy for the children in this study.
The post-instruction tonal aptitude mean scores were well above the mean
reported in the test manual (Gordon, 1986). In fact, the lower split-halves
reliability of the post-instruction administration for children in ﬁrst grade may have
been a result of the test being too easy. Another explanation for the results may
be that the children’s developmental tonal aptitude score increased as a result of
age. It is also possible that song instruction with root melody accompaniment
has no effect on developmental tonal aptitude.

Also, it is possibly a result of the children in this study having received two
years of informal music guidance prior to the study, resulting in levels of aptitude
that are so high there is no room for growth. The high aptitude levels found in
this study may be, in part, the result of the rich and varied listening vocabulary
children gained as a result of this informal guidance. Another aspect of informal
guidance and formal music instruction that may have supported the
developmental tonal aptitudes of the children in the present study is

individualized instruction. Knowing the tonal strengths and weakness of

148

individual children and individualizing instruction to meet these needs helps
children to realize their music potential.

Parental involvement in music was extremely high at this particular school
and participation in music activities was encouraged and supported through the
home and school environment. This, too may have resulted in unusually high
aptitude scores.

Another explanation for this ﬁnding is that children cannot attend to two
voices at one time — simpler may be better. If this were the case, song
instruction with root melody accompaniment would not have had an effect on
developmental tonal aptitude.

In the analysis of the pre-instruction and post-instruction IMMA tonal
aptitude scores by age-level, it was found that children in kindergarten scored
signiﬁcantly lower than the children in ﬁrst grade. Because older children score
better than younger children on IMMA as a result of maturation (Gordon, 1986),
the overall difference between the two groups in pre- and post-instruction IMMA
mean scores probably is attributable to the age difference between the two
groups. It is also possible that the difference in the amount of instruction
between the two grade levels may have contributed to this ﬁnding. Children in
kindergarten received one hour less of instruction per cycle than children in ﬁrst
grade. Also, children in ﬁrst grade had one full year more of formal music
instruction than children in kindergarten.

Song instruction with root melody accompaniment did not have an effect

on the tonal achievement of children in kindergarten and ﬁrst-grade as a result of

149

treatment. Results indicated that no signiﬁcant difference existed between the
experimental and control groups for the 1) major, minor, and composite tonality
scores, 2) major, minor, and composite keyality scores, 3) major and minor
criterion song composite scores, and 4) the tonal achievement composite scores.
This ﬁnding supports the results of Atterbury & Silcox (1993), Petzold (1966), and
Stauffer (1986) that harmonic accompaniment and its use in instruction does not
have an affect on children's singing accuracy. Longer instructional periods allow
children to achieve closer to their potential. Perhaps 25 weeks of song
instruction with root melody accompaniment was not enough to affect children’s
ability to sing with accuracy.

Another possible reason song instruction with the addition of root melody
accompaniment did not have an effect on children’s vocal achievement may have
to do with developmental maturity. Children’s ability to sing accurately may still
be in the process of developing. If this is the case, the effects of song instruction
with root melody accompaniment will not be obvious until later in the
developmental process.

The minor keyality score differences approached signiﬁcance as a result
of treatment. Children in the experimental group tended to receive higher ratings
than children in the control group. Furthermore, the kindergarten experimental
group tended to receive higher mean scores than the kindergarten control group.
The same trend also occurred between the ﬁrst-grade experimental group and
the ﬁrst-grade control group. Perhaps with a more extended study, the

differences would have been significant.

150

From the analysis of the major, minor, and composite tonality scores and
major keyality scores, it was found that children in kindergarten scored
signiﬁcantly lower than the children in first-grade. A factor that could be
contributing to the difference is that older children score better than younger
children by the advantage of age alone. Also, as before, the difference in the
amount of instruction received by the two grade levels may have contributed to
this ﬁnding.

It was found that no difference exists between children in kindergarten and
ﬁrst-grade children for the 1) major, minor, and composite keyality scores, 2)
major and minor criterion song composite scores, and 3) tonal achievement
composite scores. Children in kindergarten received the same mean scores as
children in first grade - surprisingly, age appears not to be a factor. A possible
explanation for the keyality mean scores may be that it is difficult for young
children to remain in the given keyality of a song. They can audiate tonality but
may need more experience before developing the ability to audiate keyality.
Many of the children, when performing the criterion songs, were able to begin the
songs in the given keyality but then strayed from it during the B section. Some
children were able to regain the keyality again during the ﬁnal A section of the
songs. Although Gordon (1997a, 1997c) states that children develop the ability
to sing within the keyality of songs before they accurately sing within the tonality
of songs, that did not appear to be the case in this study.

Another explanation for the similarity of keyality, criterion song, and tonal

achievement mean scores between grade levels may be that the keyality

151

dimension of the rating scale needs reﬁnement, as it had the lowest reliabilities.
This dimension of the rating scale may have been difﬁcult for the judges and
therefore did not serve as an adequate measure of the children's ability. One
bad dimension will affect the composite score. Perhaps the keyality dimension
needs modiﬁcation for future use.

Another possibility for the similarity of keyality scores between grade
levels is that there were different instructors for one of the grade levels. One
teacher provided instruction for the children in all of the kindergarten classes and
two of the ﬁrst-grade classes. Two other teachers provided instruction for the
children in one of the remaining two ﬁrst-grade classes. Because no two
teachers can teach exactly alike, the difference in instruction, as it related to
keyality, may have been a factor. Therefore, no keyality performance difference
between grade levels was found.

Song instruction with root melody accompaniment did have an effect on
the vocal improvisations of children in kindergarten and ﬁrst-grade. Children in
the experimental group received signiﬁcantly higher improvisation scores than
children in the control group. Implied harmonic functions within the original
keyality and tonality occurred most often in the melodic improvisations of children
in the experimental group. It is possible that children need exposure to the
sound of harmonic progressions within the context of familiar songs in order to
build a harmonic "speaking" vocabulary. Listening to the bass line helped
children develop a melodic sense of how melodies combine to create harmony,

and how tones function within the harmony. Tonal improvisation is most

152

meaningfully accomplished when the improvisor is audiating a harmonic context
(Gordon, 1997b 8 1998b). In addition, as would be expected, children in
kindergarten received signiﬁcantly lower improvisation scores than children in the

ﬁrst-g rade.

Conclusions and Implications for Music Education

Several conclusions about the use of root melody accompaniment with
song instruction can be reached as a result of this study. Results relating to the
effect of song instruction with root melody accompaniment on the developmental
tonal aptitude of children in kindergarten and ﬁrst grade revealed only trends in
the way song instruction with root melody accompaniment inﬂuences
developmental tonal aptitude. The results were non-signiﬁcant, and therefore,
further research is needed to verify them so that conclusive ﬁndings are
established and generalization is possible.

Trends found in this study suggest that song instruction including root
melody accompaniment may support developmental tonal aptitude more than
song instruction without root melody accompaniment. Understanding the
organization and functions of melody may occur at deeper levels when the bass
lines to songs are present. Teachers should consider presenting a portion of rote
songs with root melody accompaniment.

Trends found in this study suggest that song instruction including root
melody accompaniment may support children’s tonal achievement more than

song instruction without root melody accompaniment. A teacher who models

153

accurate singing may have more inﬂuence on children's singing accuracy when
using root melody accompaniments with song instruction.

An important observation was made with regard to instruments used to
accompany a child’s singing. The piano, soprano metalophone, soprano
xylophone, alto metalophone, alto xylophone, and tape-recorded voice worked
well for the root melody accompaniments. However, this researcher and the
other teacher of the treatment group noticed that, when the glockenspiel was
used as an accompanying instrument, the children became confused. They
seemed unclear as to whether they should sing the root melody or whether they
should sing the song. This did not occur with the other instruments or for the
tape-recorded voice. Because it caused children some difficulty, it was decided
that the glockenspiel would not be used for the remainder of the study, and
perhaps teachers should avoid using the instrument in this way as well.

Regarding the effect of song instruction with root melody accompaniment
on tonal improvisations, children who have had song instruction with root melody
accompaniment improvise better than those children who have not had such
instruction. Because of this, it seems important for teachers to include at least
some song instruction with root melody accompaniment if they are interested in

helping children become better improvisers.

Recommendations for Future Research

Results of the present study suggested that song instruction with root

melody accompaniment may facilitate the development of tonal aptitude more

154

than song instruction without root melody accompaniment. Although this was a
consistent tendency, results did not reach signiﬁcance. Further investigations of
the effect of song instruction with root melody accompaniment on developmental
tonal aptitude might include the following: (1) a more appropriate aptitude test for
children who have outgrown IMMA; (2) a different population of children with
whom IMMA (or PMMA) is the appropriate aptitude test; (3) different age-groups
to provide insight on issues relating to maturation; (4) use of an harmonic
aptitude test as a grouping and independent variable; (5) one teacher for all of
the children in the study or different teachers with equal numbers of children;
and/or (6) a longitudinal study to determine long term affects of song instruction
with root melody accompaniment on developmental tonal aptitude.

The present study has shown that the instructional treatment tended to
increase children's tonal accuracy. Song instruction with root melody
accompaniment may be a matter of developmental maturity. For the children in
this study, the ability to sing accurately may be in the process of developing. If
this is the case, the affects of root melody accompaniment with song instruction
will not be obvious until later in the developmental process. A longer treatment
period may allow singing accuracy to develop over time, possibly showing
different results than the ones found in this study. Future research might include
a longitudinal study of a group of children or perhaps a cross-sectional study with
a wider span of age-groups.

Results regarding children’s ability to sing in the given keyality showed

that maintaining keyality was difﬁcult for children in kindergarten as well as

155

children in ﬁrst grade. Gordon (1997a, 1997c) states that children develop the
ability to sing within the keyality of songs before they sing accurately within the
tonality of songs. The present study found the opposite. More research is
needed so that a better understanding of children’s ability to accurately sing the
keyality of a song can be obtained. Future research might also include a revision
of the keyality dimension of the Tonal Rating Scale. Atterbury and Silcox (1993)
suggest that the narrow range of their rating scale might have accounted for the
lack of significant differences in their study and recommend the use of a more
speciﬁc measure. The keyality dimension may not have been an accurate
measure, which ultimately would have affected the remaining criterion song
composite and tonal achievement scores.

Results of the present study showed that song instruction with root melody
accompaniment positively affects the improvisations of young children. As a
result of treatment, children are more able to perform improvisations with
harmonic functions that underpin the melody while maintaining keyality and
tonality throughout. Future research could move in several directions. Certain
harmonic progressions may be more beneﬁcial to children’s developing
improvisation skills than others. It would be advantageous for music educators to
investigate the type and sequence of root melody accompaniments most helpful
in cultivating children’s improvisational ability. Should one begin with root
melodies consisting solely of tonic and dominant chord functions and then

gradually add other chord roots to the progression? Is tonic and dominant the

156

best harmonic progression to use initially or would tonic and subdominant be a
better choice?

Alvarez (1980a & 1980b) found that seventh-grade, eighth-grade, and
college students identiﬁed harmonic progressions better with the scalar harmonic
technique than with the root melody technique. Would young children improvise
better with song instruction that included scalar harmonic accompaniment or
song instruction that included root melody accompaniment?

Gordon (1998) found that the Harmonic Improvisation Readiness Record
and the Rhythm Improvisation Readiness Record was evidence of harmonic
aptitude. Future research might investigate if improvisation achievement has to
do more with song instruction that includes root melody accompaniment or has it
more to do with harmonic improvisation aptitude?

Future research might also include a longitudinal study of the effect of
song instruction with root melody accompaniment on children’s improvisations.
Informal guidance is a time for children to build a rich listening vocabulary. How
would a large listening vocabulary of bass lines affect children’s future
improvisations?

Improvisation is a window into a child’s understanding of music. The
importance of improvisation as a valuable music skill should be more than a
notation in the National Standards for Arts Education (1994). Music educators
should ( 1) seek out information to help themselves develop improvisational skills,
(2) ﬁnd the most productive means possible to help children become competent

improvisers, and (3) actively pursue ways to make improvisation a vital part of

157

the curriculum so that children leave music programs with the skills they need to
create music of their own. Using root melody accompaniment with song
instruction is one way to facilitate the above. Hearing the bass lines to a variety

of songs is an important component in the development of improvisational skills.

158

APPENDICES

159

APPENDIX A

 

UCRIHS Permission
MICHIGAN STATE
U N I v E R s I T Y

September 4, 2001

TO: Cynthia TAGGART
204 Music Bldg.

RE: IRBS 01-570 CATEGORY: EXPEDITED 2-F
APPROVAL DATE: September 4, 2001

TITLE: THE EFFECTS OF HARMONIC ACCOMPANIMENT ON THE
DEVELOPMENTAL TONAL APTITUDE AND TONAL ACHIEVMENT OF
KINDERGARTEN AND FIRST GRADE STUDENTS

The University Committee on Research Involving Human Subjects' (UCRIHS) review of this
projectlscompleteandlampleasedtoadvisematthe rights andwelfareofthehuman
subjectsappeabbeadequatdyproteaedmdmmodstoobmlnbnnedmnsentare
appropriate. Therefore, the UCRIHS approved this project.

RENEWALS: UCRIHSepproval lsvalldforonecalendaryea'. beginningwiththeapproval
dateshownabove. Projectscontlmlngbeyondoneyearmustberenewedwiththegreen
renewal form. A maxhnum of four such expedited renewals possible. Investigators wishing to
continueaprojedbeyondthattimeneedtosubmititagainforacompleterevlew.

REVISIONS: UCRIHS must review any changes in procedures involving human subjects. prior
tolnitlatlonofthechange. lfthlsisdone atthetimeofrenewal, pleaseusathegreen renewal
form. Torevlseanapprovedprotocolatanyothertirnedurtngtheyear,sendyourwrltten
request to the UCRIHS Chair, requesting revised approval and referencing the project's IRBII
andtltle. lncludehyourrequestadescrtptlonofthechangeandanyrevbedlnstruments.
consent forms or advertisements that are applicable.

PROBLEMSICHANGES: Should either of the following arise during the course of the work.
notify UCRIHS promptly: 1) problems (unexpected side effects. complaints, etc.) involving
Mmmsubjectsor2)dtmgeehmemseaohenvkmmmtornewhfonnaﬁonhdlcaﬁng
greamrrhkbmehmnwbjedsmmeﬁstedwhmmeprotocdwaspmbuslymbwedand
approved. ’

If we can be of further assistance, please contact us at (517) 355-2180 or via email:

UCRIHSansu.edu. Please note thatall UCRIHS forms are located on theweb: .
http:/Iwww.msu.eduluserlucrlhs

 

 

AK: bd
001 Denise Guilbault
3060 Stratford St.
ammﬂmm am Misasm
mam

160

APPENDIX B

Letter to Parents

September 4, 2001

Dear Parents,

The purpose of this letter is to ask permission for your child to participate in my
dissertation on "An Investigation of the Effect of Harmonic Accompaniment on
the Developmental Tonal Aptitude, Tonal Achievement, and Improvisations of
Children in Kindergarten and First Grade." This dissertation is a requirement for
the degree of Doctor of Philosophy that I am currently pursuing at Michigan State
University.

In order to improve formal music instruction, the purpose of this study is to
examine the effect of the use of harmonic accompaniment on the developmental
tonal aptitude, tonal achievement, and improvisations of children in kindergarten
and ﬁrst grade. Speciﬁcally, this research will focus on the following: 1) Does the
addition of a root melody accompaniment to songs affect the developmental tonal
aptitude of children in kindergarten and ﬁrst grade? 2) Does the addition of a
root melody accompaniment to songs affect the song performance accuracy of
children in kindergarten and ﬁrst grade? 3) Does the addition of a root melody
accompaniment to songs affect the tonal improvisations of children in
kindergarten and ﬁrst grade?

The study will be conducted throughout this school year during your child’s
regularly scheduled music class. Two procedures will be essential for this study.
First, the Intermediate Measures of Music Audiation will be administered as a
pretest and posttest measure of the children’s music aptitude. The children are
quite familiar with this test as this is a regular procedure that is used to keep
track of their progress. Second, it will be necessary to audiotape your child’s
singing in order to accurately record his or her tonal achievement. The audio
taping will occur at the end of the study in April. The audiotapes will also be used
for analysis by three independent judges.

All aspects of your child’s involvement will be kept conﬁdential. The only persons
viewing the tapes will be the judges and myself. The judges will know the
children only by an assigned number; they will not know your child's name.
Participation is entirely voluntary, and conﬁdentiality of participants will be
protected. You can indicate your approval by signing and returning the attached
consent form. Once you have given written permission, the study will be
explained to your child. If you have any questions about this study, please call

161

me at (248) 645-3507. You may also contact the Chair of the University
Committee on Research Involving Human Subjects (UCRIHS) Dr. Ashir Kumar
at (517) 355-2180 for questions about participants' rights as human subjects of
research.

Please know that your child will receive excellent instruction during the study and
will grow musically as the year progresses. Classes will run as usual with the
exception that two kindergarten and two ﬁrst grade classes will have the addition
of harmonic accompaniments to the songs they sing. Through this research, I
hope to gain some insight into factors that may facilitate the tonal development of
young children and to improve my ability to guide children toward a deep
understanding of music through audiation.

Sincerely,

Denise Guilbault
Early Childhood Music Development Specialist
Cranbrook Schools - Brookside

162

APPENDIX C

Parental Consent Form

Parental Consent Form

Please return this form to your child’s teacher by Wednesday, September 12,
2001 if you consent to your child’s participation in this study.

Before signing, please read the attached page describing "An Investigation of the
Effects of Harmonic Accompaniment on the Developmental Tonal Aptitude, Tonal
Achievement, and Improvisations of Kindergarten and First Grade Children."
Your child's participation in this study is voluntary and will remain conﬁdential.
Verbal assent will be obtained from your child as a precondition of participating in
this study, and your child’s name will not appear in any report of results.

Students who do not choose to participate will not be penalized and they may
choose to stop their participation at any time without penalty. When this study is
completed, the overall results of it will be made available to you upon your written
request

Within these restrictions, I agree to allow my child to participate in this research
study.

Child’s Name:

 

Child’s Birth Date:

 

Parent or Legal Guardian’s Signature:

 

Today’s Date:

 

163

APPENDIX D

Song Repertoire

Maior/Duple

 

America The Beautiful .................... JRI—BK 1
Billy Billy .................................................... O
Brookside, Our School ............................... 0
Circle Round the Zero ............................... O
Cotton Songs ............................................. O
Criterion Song #1 ............................ Guilbault
Hayride ...................................................... 0
Here Comes Sally ...................................... O
Jingle Bells ................................................ O
Love Somebody .............................. JRI-BK 1
Orff Instrument Rondo ............................... O
Paw Paw Patch ......................................... O
Thanksgiving ............................................. 0
Yellow Brick Road ..................................... O
Maior/Triple

America .......................................... JRI-BK 2
Circle Around .................................. JRI-BK 2
0, How Lovely Is the Evening ........ JRI-BK 2
Maior/Unusual

Top of My Head .............................. JRI-BK 1
Harmonic Minor/Duple

Boo ............................................................ O
Criterion Song #2 ............................ Guilbault

_l-Ermonic Minor/Tripl_e

 

Mother Sent Me .............................. JRI-BK 1
Scandinavian Folk Song ................. JRI-BK 2
Temptation ................................................. O

Dorian/Triple
Matthy Groves/Little Musgrave ....... JRI-BK 1

The Lumberjack Song .................... JRI-BK 2

PhggianlDuple
The Grasshopper & Elephant ......... JRI-BK 1

PhggianlMultimetric
Song of the Pirates ......................... JRI-BK 1

Lydian/Duple
In the Mom ..................................... Guilbault

164

Lydian/Triple

 

Hello to You .................................... Guilbault
Peas ............................................... JRI-BK 2
Mixolydianl'l'riple

The Robin’s Last Will ..................... JRI-BK 2

MixolydianlUnusual
Hello Everyone ............................... Guilbault

Aeolian/Bugle
Follow the Drinkin' Gourd .......................... O

Hag Asif/Harvest Song .............................. O

Aeolian/Triple

Skin and Bones .............................. JRI-BK 1
Aeolian/Unusual
Dance from Tsakonia ..................... JRI-BK 2

MultitonallDuple
Swedish Folk Song ........................ JRI-BK 2

JRI - BK 1

Taggart, 0., Bolton, 3., Reynolds, A.,
Valerio, W, 8 Gordon, E.E. (2000).
Jump right in: The music curriculum.
Chicago: GIA Publications.

JRI - BK 2

Bolton, B., Taggart, C., Reynolds, A.,
Valerio, W, 8 Gordon, E.E. (2001).
Jump right in: The music curriculum.
Chicago: GIA Publications.

O
Other Sources

APPENDIX E

Sample of Senior Kindergarten Lesson Plans

Lesson Plan 3B

1) Hello Everyone (Mixolydian/Unusual)
2) Vocal Warm-Up
3) LSA — Tonal Unit 1-B-1

4) The Lumber Camp Song (Dorian/Triple)
a treatment group with root melody accompaniment (bass xylophone)
0 without text
a four beat pattern with rhythm sticks — tap on ﬂoor, tap together, ﬂick
twice in the air
0 sing resting tone (Re)

5) In The Mom (Lydian/Duple)

treatment group with root melody accompaniment (piano)
without text

individual bean bag toss to macrobeats and microbeats
group/individuals sing

6) Rolling (Triple Meter)

with movement

count # of times they hear ﬁrst pattern

ﬁnd part that is different from ﬁrst pattern
individuals improvise during middle section

7) Hag Asif (Aeolian/Duple)

treatment group with root melody accompaniment (piano)
without text

rote song procedure

CFM during A section; macrobeats during B section
Group/individuals sing

8) Goodbye Everyone (Mixolydian/Unusual)

165

Lesson Plan 10B

1) Hello Everyone (Mixolydian/Unusual)
2) LSA — Rhythm Unit 2-A-1
3) Vocal Wann-Up

4) Swedish Folk Song (Multitonal/Duple)

0 treatment group with root melody accompaniment (alto
xylophone)
without text
hold hands horizontally and alternate top hands to macrobeats
repeat using microbeats
group sings

5) Mother Sent Me (Harmonic Minor/Triple)
. treatment group with root melody accompaniment (alto
xylophone)
. without text
a group sings song - stop at last two measures
0 ask individual children to improvise a new ending to the song
(last two measures)

6) O, How Lovely is the Evening (Major/Triple)
0 treatment group with root melody accompaniment (alto
xylophone)
- without text
0 rote song procedure

7) Circle Round the Zero (Major/Duple)
- dance
e With text

8) Goodbye Everyone (Mixolydian/Unusual)

166

Lesson Plan 15B

1) Hello Everyone (Mixolydian/Unusual)
2) Vocal Warm-Up
3) LSA - Tonal Unit 3-B-1

4) Criterion Song #2 (Harmonic Minor/Duple)
- treatment group with root melody accompaniment (bass
metallophone)
without text
identify tonality
find # of tonic functions (La Mi Do)
teacher sings song - children audiate/sing the tonic function
patterns

5) Peas (Lydian/Triple)
0 treatment group with root melody accompaniment
(bass metallophone)
. without text
0 bilateral/alternating movements to macrobeat using rhythm
sﬁcks

6) Criterion Song #1 (Major/Duple)

. treatment group with root melody accompaniment (bass
metallophone)

- with text

. count # of dominant functions (Ti So Fa Re)

0 teacher sings song - children audiate/sing the dominant function
pattern

0 individuals sing

7) Love Somebody (Major/Duple)
. treatment group with root melody accompaniment (bass
xylophone)
without text
ask individuals to improvise tonic pattern

8) Goodbye Everyone (Mixolydian/Unusual)

167

APPENDIX F

Sample of First-Grade Lesson Plans

Lesson Plan 3B
1) Hello Everyone (Mixolydian/Unusual)
2) LSA - Rhythm Unit 3-B-1
3) Vocal Warm-Up

4) America (Major/Triple)
. treatment group with root melody accompaniment (bass
metallophone)
. with text/rote song procedure

5) Hayride (Major/Duple)
0 treatment group with root melody accompaniment (bass
metallophone)
with text

on macrobeats — tap ground in front then ground on front of person
to the right

identify beginning pitch (Mi) and last pitch (Do)
identify meter

6) The Lumber Camp Song (Dorian/Triple)
0 treatment group with root melody accompaniment (bass
metallophone)
without text
four beat pattern with rhythm sticks - tap on ﬂoor, tap together, ﬂick
twice in the air

0 ask individual children to improvise new tonic function ending

7) Hag Asif (Aeolian/Duple)
. treatment group with root melody accompaniment(bass
metallophone)
. with text

8) Brookside Our School (Major/Duple)
. treatment with bass metallophone
o with text

9) Goodbye Everyone (Mixolydian/Unusual)

168

Lesson Plan 10B

1) Hello Everyone (Mixolydian/Unusual)
2) LSA — Rhythm Unit 4-A-1
3) Vocal Warm-Up

4) Swedish Folk Song (Multitonal/Duple)

0 treatment group with root melody accompaniment (alto
xylophone)
without text
hold hands horizontally and alternate top hands to macrobeats
repeat using microbeats
group sings

5) Mother Sent Me (Harmonic Minor/Triple)
a treatment group with root melody accompaniment (alto
xylophone)
e without text
. group sings song - stop at last two measures

. ask individual children to improvise a new ending to the song
(last two measures)

6) O, How Lovely is the Evening (Major/Triple)
. treatment group with root melody accompaniment (alto
xylophone)
without text
children sing resting tone ostinato on macrobeat

half of the group sings ostinato while the other half of the group
sings the song

7) Billy, Billy (Major/Duple)
0 dance
. with text

8) Goodbye Everyone (Mixolydian/Unusual)

169

 

Lesson Plan 15B

1) Hello Everyone (Mixolydian/Unusual)
2) LSA — Rhythm Unit 4—A-1
3) Vocal Warm-Up

4) Love Somebody (Major/Duple)
0 treatment group with root melody accompaniment (bass
metallophone)
. without text
0 ask individuals to improvise dominant pattern

5) Peas (Lydian/Triple)
. treatment group with root melody accompaniment (bass
metallophone)
without text
create bilateral/alternating movements to macrobeat using
rhythm sticks
- individuals sing

6) Criterion Song #2 (Harmonic Minor/Duple)
0 treatment group with root melody accompaniment (bass
metallophone)
without text
identify tonality
ﬁnd # of tonic functions (La Mi Do)
teacher sings song - children audiate/sing the tonic function
patterns

7) Cotton Songs (Major/Duple)
‘ 0 treatment group with root melody accompaniment (bass
metallophone)
0 without text
0 rote song procedure

8) Goodbye Everyone (Mixolydian/Unusual)

170

APPENDIX G

Verbal Assent Script

“I asked you here to sing two songs that we learned in class. I will also ask you
to make up your own ending to a song. Your singing will be recorded onto this
tape recorder. Before we can begin, I need to know if this is something you
would like to do. I will give you a moment to think about it. When you are ready,

say "yes” if you would like to sing the songs or say “no” if you would not like to
sing the songs.“

171

APPENDIX H

Improvisation Antecedent Phrase

Denise Guilbault
| V

 

 

 

‘
tr
ELM. <
11:;
IL__
‘137
E11

 

 

 

 

 

 

 

I IV V '
D In .R I
' I l] ‘ r
:P a
V VI :i'l—IG

172

APPENDIX |

Criterion Songs

Criterion Song #1 - Major Tonality

Denise Guilbault

 

 

 

 

 

 

 

 

 

 

I IV W I I IV V7 I
A}. l H r: __ ﬂ 1 T‘ 1 [HI 1 1—‘1 J
T I r7 I 1 I 1 I v} a IF I L! I 1 ﬁ J
m H r ' H v a
5 V7 I V7 I I IV V7 I

 

 

 

 

 

 

Denise Guilbault
I IV V7 I I IV w 1

    

 

173

APPENDIX J

Preparatory Sequences

Major Tonality

I
£‘Ii— I 1

II I f

 

 

 

 

 

 

Harmonic Minor Tonality

' I

 

I

 

 

 

 

 

 

 

 

 

 

 

 

174

 

 

APPENDIX K

Tonal subtest of Intermediate Measures of Music Audiation Test Form

Ll

 

 

 

 

 

0‘
Q
QM
QZSEQ
@QQ
GEO
@QWO

   

..

 

 

 

 

 

 

 

 

 

 

as
I.)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.0
{530 *3. i3. {3. {4}.
Fr a / as /
3&3. @. €3© £3. £3.-
13?“ if) we; &
.Q
333. €30 $30 {3. $3.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Copyright © 1982 by GIA Publications, Inc, Chicago, Illinois.
All rights reserved.
Used with permission.

175

APPENDIX L

Tonal Achievement Rating Scale

Dimension 1: Tonality

5.

The child sings most tonic, dominant, subdominant or cadential patterns in
tune.

The child sings some tonic, dominant, subdominant or cadential patterns
Intune.

The child tonicizes one pitch, and sings intervals away from that pitch with
correct melodic direction.

The child sings small deviations around a pitch center.

The child uses a speaking voice.

Dimension 2: Keyality

5.

4.

Keyality is maintained throughout the entire song.

Keyality is maintained throughout the A, ﬁrst A’, and one of the two
remaining sections.

Keyality is maintained throughout the A and ﬁrst A’ sections.
Keyality is maintained throughout the A section.

Keyality is immediately lost.

176

APPENDIX M

Improvisation Rating Scale

Dimension: Tonal Strength

5.

4.

Maintains keyality and tonality throughout the entire song.

Harmonic functions underpin melody; keyality and tonality are maintained
throughout most but not all of the song.

Clarity of a tonal center wavers on an occasional section of the song and
harmonic functions are unclear.

A tonal center is clear only occasionally.
No sense of a tonal center at any point in the song.

The child does not improvise.

177

REFERENCES

178

REFERENCES

Adderly, C. L., III (1999). Preparation of future band directors relative to the
National Standards-Goals 2000. Journal of Band Research, 35(1), 63-73.

Alvarez, M. (1980a). An experimental study of the effect of scalar and
fundamental bass aural perception techniques on teaching seventh- and
eight-grade general music students to identify primary harmonic functions. .
Unpublished master’s thesis, State University of New York at Buffalo. @-

Alvarez, M. (1980b). The effects of classifying, sequencing, and coding on the I’
identiﬁcation of harmonic functions (Doctoral dissertation, State University ‘
of New York at Buffalo, 1980). Dissertation Abstracts International,
41(06A), 2490.

Amchin, R. A. (1995). Creative musical response: The effects of teacher-student
interaction on the improvisation abilities of fourth- and ﬁfth-grade students
(Doctoral dissertation, University of Michigan, 1995). Dissertation
Abstracts International, 56(08A), 3044.

Andress, B. (1986). Toward an integrated developmental theory for early
Childhood music education. Bulletin of the Council for Research in Music
Education, 86, 10-18.

Apfelstadt, H. (1984). Effects of melodic perception instruction on pitch
discrimination and vocal accuracy of kindergarten children. Journal of
Research in Music Education, 32(1), 15-24.

Atterbury, B. W. (1984a). Are you really teaching children how to sing? Music
Educators Journal, 70(8), 43-45.

Atterbury, B. W. (1984b). Children's singing voices: A review of selected
research. Council for Research in Music Education, 80, 51-63.

Atterbury, B. W. & Silcox, L. (1993). The effect of piano accompaniment on
kindergarteners’ developmental singing ability. Journal of Research in
Music Education, 41 (1 ), 40-47.

Azzara, C. D. (1992). The effect of audiation-based improvisation techniques on
the music achievement of elementary instrumental music students
(Doctoral dissertation, The University of Rochester, Eastman School of
Music, 1992). Dissertation Abstracts International, 53(04A), 1088.

179

Azzara, C. D., Grunow, R. F., 8 Gordon, E. E. (19973). Creativity in
improvisation: Book 1. Chicago: G.I.A. Publications.

Azzara, C. D., Grunow, R. F., 8 Gordon, E. E. (1997b). Creativity in
improvisation: Book 2. Chicago: G.I.A. Publications.

Azzara, C. D. (1999). An aural approach to improvisation. Music Educators
Journal, 86(3), 21-5.

Begley, S. (1996). Your child’s brain. Newsweek, 127(8), 54-58.

Bentley, A. (1968). Monotones: A comparison with “normal” singers in terms of
incidence and musical abilities. Music Education Research Papers No. 1.
London: Novello.

Bentley, A. (1969). Measurements and development of musical abilities: Some
research interests and ﬁndings. Joumal of Research in Music Education,
1 7, 41 ~46.

Berke, M. K. (2000). The ability of preschool children to recognize chord
changes and audiate implied harmony (Doctoral dissertation, The
University of Arizona, 2000). Dissertation Abstracts lntemational,
61(05A), 1776.

Bertaux, B. (1989). Teaching children of all ages to use the singing voice, and
how to work with out of tune singers. In D. L. Walters 8 C. C. Taggart
(Eds), Readings in music Ieaming theory (pp. 92-104). Chicago: G.I.A.
Publications.

Boardman, E. L. (1968). An investigation of the effect of preschool training on
the development of vocal accuracy in young children (Doctoral
dissertation, University of lllinios, 1964). Dissertation Abstracts
lntemational, 25(02), 1245.

Bolton, B., Taggart, C., Reynolds, A., Valerio, W, 8 Gordon, E. E. (2001). Jump
right in: The music curriculum. Chicago: GIA Publications.

Bridges, V. A. (1966). An exploratory study of the harmonic discrimination ability
of children in kindergarten through grade three in two selected school
(Doctoral dissertation, Ohio State University, 1965). Dissertation
Abstracts lntemational, 26(07), 3692.

Briggs, N. L. (1987). Creative improvisation: A musical dialogue (Doctoral
dissertation, University of California, San Diego, 1986). Dissertation
Abstracts lntemational, 4 7(08A), 2787.

180

O Briscuso, J. J. (1972). A study of ability in spontaneous and prepared jazz
improvisation among students who possess different levels of musical
aptitude (Doctoral dissertation, University of Iowa, 1972). Dissertation
Abstracts lntemational, 33(04A), 1761.

Brophy, T. S. (1999). The melodic improvisations of children ages six through
twelve: A developmental perspective (Doctoral dissertation, University of
Kentucky, 1998). Dissertation Abstracts lntemational, 59(09A), 3386.

Brown, C. J. (1988). The effect of two assessment procedures on the range of
children’s singing voices. Unpublished master’s thesis, Indiana University,
Bloomington.

Cassidy, J. W. (1993). Effects of various sight singing strategies on nonmusic
majors’ pitch accuracy. Journal of Research in Music Education, 41(4),
293-302.

Clayton, L. S. (1986). An investigation of the effect of a simultaneous pitch
stimulus on vocal pitch accuracy. Unpublished master’s thesis, Indiana
University, Bloomington.

Clegg, B. W. (1966). A comparative study of primary grade children ’s ability to
match tones. Unpublished master’s thesis, Brigham Young University,
Provo, Utah.

Cohen, V. (1981). The emergence of musical gestures in kindergarten children
(Doctoral dissertation, University of Illinois at Urbana-Champaign, 1980).
Dissertation Abstracts International, 41(11A), 4637.

Collins, I. H. (1981). The child voice. Choristers Guild Letters, 23(4), 63-65.

Consortium of National Arts Education Associations (1994). National standards
for arts education. Reston, VA: Music Educators National Conference.

Cooper, N. A. (1995). Children’s singing accuracy as a function of grade level,
gender, and individual versus unison singing. Journal of Research in
Music Education, 43(3), 222-231.

Costa-Giomi, E. (1992). Recognition of chord changes by four- and ﬁve-year-old
American and Argentine children (Doctoral dissertation, Ohio State
University, 1991). Dissertation Abstracts lntemational, 52(08A), 2749.

Davies, A. D. M., 8 Roberts, E. (1975). Poor pitch singing: A survey of its
incidence in school children. Psychology of Music, 3(2), 24-36.

181

Davies, J. B., 8 Yelland, A. (1977). Effects of two training procedures on the
production of melodic contour in short-term memory for tonal sequence.
Psychology of Music, 5(2), 3-9.

DeLorenzo, L. C. (1989). A ﬁeld study of sixth-grade students’ creative music
problem-solving processes. Journal of Research in Music Education,
37(3), 188-200.

DiBlassio, R. V. (1985). An experimental study of the development of tonal and
rhythmic capabilities of ﬁrst grade children (Doctoral dissertation, Temple
University, 1984). Dissertation Abstracts lntemational, 46(03A), 0641.

Doig, D. (1941 ). Creative music I. Music composed for a given text. Journal of
Educational Research, 35(4), 263-271.

Doig, D. (1942a). Creative music ll. Music composed on a given subject. Journal
of Educational Research, 35(5), 344-355.

Doig, D. (1942b). Creative music III. Music composed to illustrate given musical
problems. Journal of Educational Research, 36(4), 241 -253.

English, W. H. (1985). The relative effectiveness of the amount of piano
accompaniment in beginning strings class instruction (Doctoral
dissertation, University of Houston, 1985). Dissertation Abstracts
International, 46(06A), 1550.

Feierabend, J. M. (1984). The effects of speciﬁc tonal pattern training on singing-
and aural discrimination abilities of ﬁrst grade children (Doctoral
dissertation, Temple University, 1984). Dissertation Abstracts
lntemational, 45(01A), 01 10.

Feierabend, J., Saunders, T. C., Holahan, J., 8 Getnick, P. (1998). Song
recognition among preschool-age children: An investigation of words and
music. Journal of Research in Music Education, 46(3), 351-359.

Flohr, J. W. (1980). Musical improvisation behavior of young children (Doctoral
dissertation, University of Illinois at Urbana-Champaign, 1979).
Dissertation Abstracts International, 40(10A), 5355.

Flohr, J. W. (1981). Short-term music instruction and young children’s
developmental music aptitude. Journal of Research in Music Education,
29(3), 219-223.

Flohr, J. W. (1985). Young children's improvisations: Emerging creative thought.
Creative Child and Adult Quarterly, 10, 79-85.

182

Flowers, P. 8 Dunne-Sousa, D. (1990). Pitch-pattem accuracy, tonality, and
vocal range in preschool children’s singing. Journal of Research in Music
Education, 38(2), 102-114.

Freundlich, D. (1979). The development of musical thinking: Case studies in
improvisation (Doctoral dissertation, Harvard University, 1978).
Dissertation Abstracts lntemational, 39(11A), 6617.

F roehlich, H. C. (1977). An investigation of the relationship of selected
observational variables to the teaching of singing (Doctoral dissertation,
University of Texas at Austin, 1976). Dissertation Abstracts lntemational,
37(08A), 4947.

Geringer, J. M. (1983). The relationship of pitch matching and pitch
discrimination abilities of preschool and fourth grade students. Journal of
Research in Music Education, 31 (2), 93-99.

Geringer, J. M., Nelson, J. K., 8 Kostka, M. C. (1981). Differential assessments
of child and adult vocal ranges. Contributions to Music Education, 8, 39-
45.

Goetze, M. (1986). Factors affecting accuracy in children’s singing. (Doctoral
dissertation, University of Colorado, 1985). Dissertation Abstracts
International, 46(10A), 2955.

Goetze, M., 8 Horii, Y. (1989). A comparison of the pitch accuracy of group and
individual singing in young children. Council for Research in Music
Education, 99, 57-73.

Goetze, M., Cooper, N., 8 Brown, C. (1990). Recent research on singing in the
general music classroom. Council for Research in Music Education, 104,
16-37.

Gordon, E. E. (1974). Toward the development of a taxonomy of tonal and
rhythm patterns: Evidence of difﬁculty level and growth rate. Experimental
Research in the Psychology of Music: Studies in the Psychology of Music,
9, 39-232.

Gordon, E. E. (1976). Tonal and rhythm pattems: An objective analysis. Albany,
NY: State University of New York Press.

Gordon, E. E. (1979). Primary measures of music audiation. Chicago: G.I.A.
Publications.

183

Gordon, E. E. (1980a). Developmental music aptitudes among inner-city primary
children. Bulletin of the Council on Research in Music Education, 63, 25-
30.

Gordon, E. E. (1980b). The assessment of music aptitudes of very young
children. Gifted Child Quarterly, 24(3), 107-111.

Gordon, E. E. (1982). lntenrrediate measures of music audiation. Chicago:
G.I.A. Publications.

Gordon, E. E. (1986). Manual for the primary measures of music audiation and
the intermediate measures of music audiation. Chicago: G.I.A.
Publications.

Gordon, E. E. (1989a). Audiation, music learning theory, music aptitude, and
creativity. In J. W. Richmond (Ed.), The Proceedings of the Suncoast
Music Education Forum on Creativity (pp. 75-89). Tampa, FL: The
University of South Florida.

Gordon, E. E. (1989b). Audie. Chicago: G.I.A. Publications.

Gordon, E. E. 8 Woods, D. G. (1990a). Rhythm register book one from jump
right in: The music curriculum (Rev. ed.). Chicago: G.I.A. Publications.

Gordon, E. E. 8 Woods, D. G. (1990b). Tonal register book one from jump right
in: The music curriculum (Rev. ed.). Chicago: G.I.A. Publications.

Gordon, E. E. 8 Woods, D. G. (1992). Reference handbook for using Ieaming
sequence activities from jump right in: The music curriculum (2"d Rev.
ed.). Chicago: G.I.A. Publications.

Gordon, E. E. (19973). Learning sequences in music: Skill, content, and
patterns. Chicago: G.I.A. Publications.

Gordon, E. E. (1997b). Preparing young children to improvise at a later time.
Early Childhood Connections 3(4), 6-12.

Gordon, E. E. (1997c). A music Ieaming theory for newborn and young children.
Chicago: G. I. A. Publications.

Gordon, E. E. (19983). Introduction to research and the psychology of music.
Chicago: G.I.A. Publications.

Gordon, E. E. (1998b). Manual for the Harmonic improvisation readiness record
and rhythm improvisation readiness record. Chicago: G.I.A. Publications.

184

Gordon, E. E. (1999). All about audiation and music aptitudes. Music Educators
Journal, 86(2), 41-44.

Gordon, E. E. (2000). Studies in harmonic and rhythmic improvisation readiness.
Chicago: G.I.A. Publications.

Gordon, E. E. (2001). Test validity and cuniculum development: Three
longitudinal studies in music. (GIA Monograph Series). Chicago: G.I.A.
Publications.

Gould, A. O. (1969). Developing specialized programs for singing in the
elementary school. Bulletin of the Council for Research in Music
Education, 1 7, 9-22.

Gouzouasis, P. (1993). Music audiation: A comparison of the music abilities of
kindergarten children of various ethnic backgrounds. The Quarteriy
Journal of Music Teaching and Learning, 4(2), 70-76.

Green, G. A. (1987). The effect of vocal modeling on pitch-matching accuracy of
children in grades one through six (Doctoral dissertation, The Louisiana
State University and Agricultural and Mechanical College, 1987).
Dissertation Abstracts lntemational, 48(06A), 1410.

Green, G. A. (1990). The effect of vocal modeling on pitch-matching accuracy of
elementary schoolchildren. Journal of Research in Music Education,
38(3), 225-231.

Green, G. A. (1994). Unison verses individual singing and elementary students’
vocal pitch accuracy. Journal of Research in Music Education, 42(2), 105-
114.

Grutzmacher, P. A. (1987). The effect of tonal pattern training on the aural
perception, reading recognition, and melodic sight-reading achievement of
ﬁrst-year instrumental music students. Journal of Research in Music
Education, 35(3), 171-181 .

Hale, M. R. (1977). An experimental study of the comparative effectiveness of
harmonic and melodic accompaniment in singing as it relates to the
development of a sense of tonality (Doctoral dissertation, State University
of New York at Buffalo, 1976). Dissertation Abstracts lntemational,
37(08A), 4947.

Hermanson, L. W. (1972). An investigation of the effects of timbre on
simultaneous vocal pitch acuity of young children (Doctoral dissertation,
Columbia University, 1971 ). Dissertation Abstracts lntemational, 32(07A),
3558.

185

Holahan, J. M. (1983). The effects of four conditions of "same” and “different”
instruction on the developmental music aptitudes of kindergarten children
receiving tonal pattern training (Doctoral dissertation, Temple University,
1983). Dissertation Abstracts lntemational, 44(05A), 1370.

Hufstader, R. A. (1977). A learning sequence of selected music listening skills
for grades one through seven (Doctoral dissertation, University of Iowa,
1976). Dissertation Abstracts International, 37(08A), 4948.

Jarjisian, C. S. (1983). Pitch pattern instruction and the singing achievement of
young children. Psychology of Music, 11(1), 19-25.

Jersild, A. (1932). Training and growth in the development of children: A study
of the relative inﬂuence of Ieaming and maturation. New York: Teachers
College, Columbia University.

Jessup, L. L. (1984). The comparative effects of indirect and direct music
teaching upon the developmental music aptitudes and music achievement
of early primary grade students (Doctoral dissertation, Temple University,
1984). Dissertation Abstracts International, 45(06A), 1678.

Jones, M. (1971). A pilot study in the use of a vertically-arranged keyboard
instrument with the uncertain singer. Journal of Research in Music
Education, 19(2), 183-194.

Jordan-DeCarbo, J. (1982). Same different discrimination techniques, readiness
training, pattern treatment, and sex on aural discrimination and singing of
tonal patterns by kindergarteners. Journal of Research in Music
Education, 30(4), 237-246.

Joseph, A. S. (1983). A Dalcroze eurhythmics approach to music learning in
kindergarten through rhythmic movement, ear-training and improvisation.
accuracy in sight reading achieved by college elementary group piano
students (Doctoral dissertation, Carnegie-Mellon University, 1982).
Dissertation Abstracts lntemational, 44(02A), 420.

Joyner, D. R. (1969). The monotone problem. Journal of Research in Music
Education, 17(1), 114-125.

Kalmer, M 8 Balasko, G. (1987). "Musical mother tongue” and creativity in

preschool children’s melody improvisations. Bulletin for the Council of
Research in Music Education 91, 77-89.

186

 

Kane, M. J. (1994). The effects of teacher training on the developmental music
aptitude and music achievement of kindergarten students (Doctoral
dissertation, Temple University, 1994). Dissertation Abstracts
International, 55(04A), 0899.

Kelly, S. N. (1998). Preschool classroom teachers” perceptions of useful music
skills and understandings. Journal of Research in Music Education, 46(3),
374-383.

Kimble, E. P. (1984). The effect of various factors on the ability of children to
sing an added part (Doctoral dissertation, University of Georgia, 1983).
Dissertation Abstracts lntemational, 44(08A), 2400.

Kirkland, M. J. (1996). South Carolina schools and Goals 2000: National
Standards in music (Doctoral dissertation, University of South Carolina,
1996). Dissertation Abstracts lntemational, 57(03A), 1069.

Klinger, R., Campbell, P., 8 Goolsby, T. (1998). Approaches to children’s song
acquisition: Immersion and phrase-by-phrase. Journal of Research in
Music Education, 46(1), 24-34.

Kotulak, R. (1993, July 4). Discoveries show brain shapes itself with senses.
The Charlotte Observer, pp. A1, A8-A9.

Kramer, S. J. (1986). The effects of two different music programs on third and
fourth grade children’s ability to match pitches vocally (Doctoral
dissertation, Rutgers University, 1985). Dissertation Abstracts
lntemational, 46(09A), 2609.

Kratus, J. (1985). The use of melodic and rhythmic motives in the original
songs of children aged 5 to 13. Contributions to Music Education, 12,
1-8.

Kratus, J. K. (1986). Rhythm, melody, motive, and phrase characteristics of
original songs by children aged ﬁve to thirteen (Doctoral dissertation,
Northwestern University, 1985). Dissertation Abstracts lntemational,
46(11A), 3281.

Kratus, J. (1989). A time analysis of the compositional processes used by
children ages 7 to 11. Journal of Research in Music Education, 37(1), 5-
20.

Kratus, J. (1991a). Growing with improvisation. Music Educators Journal, 78(4),
35-40.

187

Kratus, J. (1991b). Orientation and intentionality as components of creative
musical activity. Perspectives in Music Education, 2, 4-8. [Reprinted from
C. Doane 8 J. Richmond (Eds.), Proceedings of the 1989 Suncoast Music
Education Forum (pp. 93-104). Westport, CT: Greenwood Press]

Kratus, (1991c). Characterization of the compositional strategies used by
children to compose a melody. Canadian Journal of Research in Music
Education, 33, 95-103.

Kratus, J. (1994). Relationships among children’s music audiation and their
compositional processes and products. Journal of Research in Music
Education, 42(2), 1-23.

Kratus, J. (1995). The effect of composing tempo on the musical characteristics
of children’s compositions. Contributions to Music Education, 22, 40-48.

Kratus, J. (1996). A developmental approach to teaching musical improvisation.
International Journal of Music Education, 26, 27-38.

Kuhn, T. L. 8 Sims, W. L. (1983). The effect of simultaneous melodic and
harmonic accompaniment, pitch level, and song tones on ﬁrst-grade
student’s ability to sing correct pitches. In P.T. Tallarico (Ed.),
Contributions to Symposium/83: The Bowling Green State University
Symposium on Music Teaching and Research (pp. 153-166). Bowling
Green, Ohio: Bowling Green State University.

Laczo’, Z. (1981). A psychological investigation of improvisational abilities in the
lower and higher classes of the elementary school. Bulletin of the Council
for Research in Music Education 66-67, 39-45.

Lange, D. (2000). The effect of the use of text in music instruction on the tonal
aptitude, tonal accuracy, and tonal understanding of kindergarten children
(Doctoral dissertation, Michigan State University, 1999). Dissertation
Abstracts lntemational, 60(10A), 3623.

Levinowitz, L. M. (1987). An experimental study of the comparative effects of
singing songs with words and without words on children in kindergarten
and ﬁrst grade (Doctoral dissertation, Temple University, 1987).
Dissertation Abstracts lntemational, 48(04A), 0863.

Levinowitz, L. (1989). An investigation of preschool children's comparative
capability to sing songs with and without words. Bulletin of the Council for
Research in Music Education, 100, 14-19.

188

Levinowitz, L., Barnes, P., Guerrini, S., Clement, M., D’April, P., 8 Morey, M.
(1998). Measuring singing voice development in the elementary general
music classroom. Journal of Research in Music Education, 46(1), 35-47.

Madsen, C. K., Wolfe, D. E., 8 Madsen, C. (1969). The effect of reinforcement
and directional scalar methodology on intonational improvement. Bulletin
of the Council for Research in Music Education, 18, 22-33.

Mallet, C. A. (2000). An examination of parent/caregiver attitudes toward music
instruction, the nature of the home musical environment, and their
relationship to the developmental music aptitude of preschool children
(Doctoral dissertation, University of Nebraska, 2000). Dissertation
Abstracts lntemational, 61 (04A), 1335.

Martin, M. A. (1993). An examination of the Orff-Schulwerk approach to music
education in a public elementary school: A replication study (Doctoral
dissertation, University of North Carolina, Chapel Hill, 1992). Dissertation
Abstracts International, 53(08A), 2728.

Mizener, C. P. (1993). Attitudes of children toward singing and choir participation
and assessed singing skill. Journal of Research in Music Education,
41(3), 233-245.

Montano, D. R. (1983). The effect of improvisation on given rhythms on rhythmic
accuracy in sight reading achieved by college elementary group piano
students (Doctoral dissertation, University of Missouri, Kansas City, 1992).
Dissertation Abstracts lntemational, 44(06A), 1720.

Moore, R. S. (1991). Comparison of children’s and adults’ vocal ranges and
preferred tessituras in singing familiar songs. Bulletin of the Council for
Research in Music Education, 107, 13-22.

Moore, R. S. (1994). Effects of age, sex, and melodic/harmonic patterns on
vocal pitch-matching skills of talented 8-11-year-olds. Journal of
Research in Music Education, 42(1), 5-13.

Moore, R. S., Brotons, M., Fyk, J. 8 Castillo, A. (1997). Effects of culture, age,
gender, and repeated trials on rote song learning skills of children 6-9
years old from England, Panama, Poland, Spain, and the United States.
Bulletin of the Councillor Research in Music Education, 133, 83—88.

Moorhead, G. 8 Pond, D. (1978). Music for young children. Santa Barbara, CA:

Pillsbury Foundation for the Advancement of Music Education. (Reprinted
from the 1941-1951 editions.)

189

Munsen, S. C. (1987). An description and analysis of an Orff-Schulwerk program
of music education (Doctoral dissertation, University of Illinois, Urbana-
Champaign, 1986). Dissertation Abstracts lntemational, 47(09A), 3351.

Nash, M. (1997). Fertile minds. Time, February 3, 49-56.

Partchey, K. C. (1974). The effects of feedback, models, and repetition on the
ability to improvise melodies (Doctoral dissertation, Pennsylvania State
University, 1973). Dissertation Abstracts lntemational, 35(01A), 240.

Pedersen, D. M., 8 Pedersen, M. O. (1970). The relationship between pitch
recognition and vocal pitch production in sixth-grade students. Joumal of
Research in Music Education, 28(3), 265-272.

Petzold, R. G. (1963). The development of auditory perception of musical
sounds by children in the ﬁrst six grades. Journal of Research in Music
Education, 11(1), 21-43.

Petzold, R. G. (1966). Auditory perceptions of musical sounds by children in the
ﬁrst six grades (Cooperative Research Project No. 1051 ). Washington,
DC: Ofﬁce of Education, US. Department of Health, Education, and
Welfare.

Phillips, K. H. (1996). Teaching kids to sing. New York: Schirmer Books.

Phillips, K. H., 8 Aitchison, R. E. (1997a). Effects of psychomotor instruction on
elementary general music students’ singing performance. Journal of
Research in Music Education, 45(2), 185-196.

Phillips, K. H., 8 Aitchison, R. E. (1997b). The relationship of singing accuracy to
pitch discrimination and tonal aptitude among third grade students.
Contributions to Music Education, 24(1), 7-22.

Prevel, M. (1979). Emergent patterning in children's musical improvisations.
Canadian Music Educator, 15, 13-15.

Regelski, T. A. (1981). Teaching general music. New York: Schirmer Books

Reinhardt, D. (1990). Preschool children's use of rhythm in improvisation.
Contributions to Music Education, 17, 7-19.

Richner, S. S. (1976). The effect of classroom and remedial methods of music
instruction on the ability of inaccurate singers, in the third, fourth, and ﬁfth
grades, to produce pitches (Doctoral dissertation, University of Idaho,
1976). Dissertation Abstracts lntemational, 37(03A), 1447.

190

Ritchie, T. V. (1962). A study of the effects of diatonic harmony upon the aural
perception of selected melodic fragments (Doctoral dissertation, Indiana
University, 1960). Dissertation Abstracts lntemational, 21 (03), 0642.

Rutkowski, J. (1986). The effect of restricted song range in kindergarten
children’s use of singing voice and developmental music aptitude
(Doctoral dissertation, State University of New York at Buffalo, 1986).
Dissertation Abstracts lntemational, 47(08A), 2072.

Rutkowski, J. (1996). The effectiveness of individual/smaIl-group singing
activities on kindergarteners’ use of singing voice and developmental
music aptitude. Journal of Research in Music Education, 44(4), 353-368.

Sherman, R. (1971). Creativity and the condition of knowing in music. Music
Educators Journal, 58(2), 18-22.

Sims, W. L., Moore, R. S., 8 Kuhn, T. L. (1982). Effects of female and male
vocal stimuli, tonal pattern length and age on vocal pitch-matching abilities
of young children from England and the United States. Psychology of
Music, Special Issue: Proceedings of the IX International Seminar on
Research in Music Education, 104-108.

Sinor, E. (1985). The singing of selected tonal patterns by preschool children
(Doctoral dissertation, Indiana University, 1984). Dissertation Abstracts
lntemational, 45(1 1A), 3299.

Smale, M. J. (1988). An investigation of pitch accuracy of four- and ﬁve-year-old
singers (Doctoral dissertation, University of Minnesota, 1987).
Dissertation Abstracts lntemational, 48(08A), 2013.

Small, A., 8 McCachern, F. L. (1983). The effect of male and female vocal
modeling on pitch-matching accuracy of ﬁrst-grade children. Joumal of
Research in Music Education, 31 (3), 227-233.

Smith, R. S. (1974). Factors related to children’s in-tune singing abilities
(Doctoral dissertation, West Virginia University, 1973). Dissertation
Abstracts lntemational, 34(11A), 7271.

Stamou, L. (1999). The effect of Suzuki instruction and early childhood music
experiences on the developmental music aptitude and performance
achievement of beginning Suzuki string students (Doctoral dissertation,
Michigan State University, 1998). Dissertation Abstracts lntemational,
59(10A), 3769.

191

Stauffer, S. L. (1986). An investigation of the effects of melodic and harmonic
context on the development of singing ability in primary grade children
(Doctoral dissertation, University of Michigan, 1985). Dissertation
Abstracts International, 46(07A), 1862.

Sterling, P. A. (1985). A developmental study of the effects of accompanying
harmonic context on children’s vocal pitch accuracy of familiar melodies
(Doctoral dissertation, University of Miami, 1984). Dissertation Abstracts
lntemational, 45(08A), 2436.

Taggart, C. C. (1989). The measurement and evaluation of music aptitudes and
achievement. In D. L. Walters 8 C. C. Taggart (Eds), Readings in music
Ieaming theory (pp. 45-53). Chicago: G. I. A. Publications.

Taggart, C. C. (1997, November). A study of developmental music aptitude.
Paper presented at the New Directions in Music Education: Early
Childhood Music Conference, Michigan State University, East Lansing, MI.

Taggart, 0., Bolton, 8., Reynolds, A., Valerio, W, 8 Gordon, E. E. (2000). Jump
right in: The music curriculum. Chicago: GIA Publications.

Thomas, R. B. (1991). Musical ﬂuency: MMCP and today’s curriculum. Music
Educators Journal, 78(4), 26-29.

Torrance, ER (1962). Guiding creative talent. Englewood Cliffs, NJ: Prentice-
Hall.

Vaughn, M. M. (1981). Intercultural studies in children’s natural singing pitch and
walking tempo. Bulletin of the Council for Research in Music Education,
66-67, 96-101.

Waddell, G. (1989). Teaching a rote song. In D. L. Walters 8 C. C. Taggart
(Eds), Readings in music Ieaming theory (pp. 84-91). Chicago: G.I.A.
Publications.

Webster, P. (1987). Reﬁnement of a measure of creative thinking in music. In
Madsen, C. 8 Prickett, C. (Editors). Applications of research in music
behavior (pp. 257-271 ). Tuscaloosa: University of Alabama Press.

Webster, P. (1989). Creative thinking in music: The assessment question. In J.
Richmond (Ed.), Proceedings of the Suncoast Music Education Forum on
Creativity (pp. 40-74). Tampa, FL: University of South Florida.

Welch, G. (1979a). Poor pitch singing: A review of the literature. Psychology of
Music, 7(1), 50-58.

192

Welch, G. (1979b). A schema theory of how children learn to sing in tune.
Psychology of Music, 7(2), 13-31 .

Welch, G. (1985). Variability of practice and knowledge of results as factors in
learning to sing in tune. Bulletin of the Council for Research in Music
Education, 85, 238-247.

Welch, G. (1986). A developmental view of children’s singing. British Journal of
Music Education, 3(3), 295-303.

Welch, G., Sergeant, D., 8 White, P. (1995/1996). The singing competencies of
ﬁve-year-old developing singers. Bulletin of the Council for Research in
Music Education, 127, 163-169.

Wig, J. A. (1981). The effect of instruction in music composition strategies on
middle school band students’ ability to improvise melodies (Doctoral
dissertation, Pennsylvania State University, 1980). Dissertation Abstracts
lntemational, 41 (10A), 3424.

Wiggins, J. (1995). Learning through creative interactions with music. General
Music Today, 8(3), 11-15.

Wilson, D. S. (1971). A study of the child voice from six to twelve. (Doctoral
dissertation, University of Oregon, 1970). Dissertation Abstracts
lntemational, 31(10A), 5453.

Yarbrough, C. (1991). Inaccurate singers: An exploratory study of variables
affecting pitch-matching. Bulletin of the Council for Research in Music
Education, 107, 23-34.

Yarbrough, C. (1992). The effect of vibrato on the pitch-matching accuracy of
certain and uncertain singers. Journal of Research in Music Education,
40(1), 30-38.

Zimmerman, M. P. (1971). Musical characteristics of young children. Reston,
VA: Music Educators National Conference.

Zwissler, R. N. (1972). An investigation of the pitch discrimination skills of ﬁrst-
grade children identiﬁed as accurate and those identiﬁed as inaccurate
singers (Doctoral dissertation, University of California, Los Angeles, 1971).
Dissertation Abstracts lntemational, 32(07A), 4056.

193

I I} III- I

. II'IIIKIIIII

‘II

.III

I? 4" Ir... I

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

llllllljlllllIllllllllllllllllllllllllllllllIllllllllllllll

1293 02428 8387