—
_———
._—__
_____

 

 

LIBRARY
1 MIChIgan State
G , 1) University

 

This is to certify that the
thesis entitled

AN INVESTIGATION OF THE USE OF SINGING GAMES
IN MEASURING SINGING VOICE DEVELOPMENT
IN YOUNG CHILDREN

presented by

LEA ANNE STEENWYK

has been accepted towards fulﬁllment
of the requirements for the

M. A. degree in School of Music

 

 

’

l

 

Major Professo 7 nature

I'ZILo IOLl—

Date

MSU is an Afﬁrmative Action/Equal Opportunity Institution

u-c-C-I—I-v-ao

   

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Acc&Pres/ClRC/DateDue.indd

 

 

AN INVESTIGATION OF THE USE OF SINGING GAMES IN MEASURING
SINGING VOICE DEVELOPMENT IN YOUNG CHILDREN

By

Lea Anne Steenwyk

A THESIS
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
MASTER OF MUSIC IN MUSIC EDUCATION
School of Music

2004

ABSTRACT

AN INVESTIGATION OF SINGING GAMES AS A MEASURE OF
SINGING VOICE DEVELOPMENT IN YOUNG CHILDREN

By
Lea Anne Steenwyk

The purpose of this research was to investigate techniques for measuring singing
voice development. The subjects were 78 third grade children from a suburban
elementary school. First, children were administered a standardized test of singing voice
development, Rutkowski’s Singing Voice Development Measure (SVDM). Then, they
played a singing game in their music class that required them to sing alone. Their solo
performances were recorded and scored by two independent judges using Rutkowski’s
rating scale. The scores on SVDM were correlated with the scores on the singing game
to determine whether singing voice development could be measured with validity in a
singing game Context. Inter-judge reliability was .84 for SVDM and .90 for the singing
game. When the scores for the two measures were correlated, a relationship of .74 was
found. This is a moderately high correlation, providing evidence that singing games can

be a valid context in which to measure children’s singing voice development.

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Cynthia Taggart, for all her help and
direction throughout the thesis process. The following thesis has been made possible
because of her. I would also like to extend my gratitude to Dr. Mitch Robinson and Dr.
Alan Gosman who served as committee members and provided much valuable insight.

I would like to thank Lincoln Elementary and Principal Tom DeGraaf for opening
their doors and supporting me in conducting research in the school. Much appreciation is
extended toward teachers Lori DeVisser, Nancy Klaver, Raynetta McGee, and Mike
VanKampen in being ﬂexible and helpful throughout the study. Thank you to Matt
Lucas and Dave Heuvelhorst for their assistance in recording the data. I would also like
to thank the third grade students at Lincoln Elementary for taking part in this study and
being fabulous to work with. Much gratitude is extended to my colleagues and friends,
Lisa Koops, Renee VandeWege, and Ruby Boyden, for their help and support.

Finally, I would like to thank my husband Eric. He is the one who believed in me
and encouraged me from the start to take on this endeavor. He has been there every step

of the way as my greatest cheerleader and friend.

iii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................. vi
LIST OF FIGURES ............................................................................... vii
CHAPTER ONE
LITERATURE REVIEW ........................................................................ 1
Singing ..................................................................................... 1
Importance of Singing ........................................................... 1
Singing Voice Development ................................................... 3
Assessing Singing ......................................................................... 6
Criterion Measures ............................................................... 6
Practical Assessment ............................................................ 11
Young Children’s Growth and Development ......................................... 15
Play ................................................................................ 15
Play verses Work ................................................................ 18
Music and Play ................................................................... 20
Games ............................................................................. 23
Singing Games ................................................................... 25
Purpose ............................................................................ 27
Problem ........................................................................... 27
CHAPTER TWO
RELATED RESEARCH ......................................................................... 28
Singing Voice Development Measure ........................................ 28
Other Studies Using SVDM ................................................... 3O
SVDM tested in a practical environment ..................................... 33
Validity ........................................................................... 35
CHAPTER THREE
METHOD .......................................................................................... 37
Subjects ............................................................................ 37
Design and Procedures ........................................................... 37
CHAPTER FOUR
RESULTS AND INTERPRETATIONS ....................................................... 42
Reliabilities ........................................................................ 42
Means and Standard Deviations ................................................ 42
Correlation as a Measure of Concurrent Validity ........................... 43
CHAPTER FIVE
CONCLUTIONS AND RECOMMENDATIONS ............................................ 44
Purpose ............................................................................ 44
Problem ........................................................................... 44

iv

Summary .......................................................................... 44
Future Research .................................................................. 45

REFERENCES .................................................................................... 47

LIST OF TABLES

TABLE ONE ....................................................................................... 43

vi

LIST OF FIGURES

FIGURE ONE ....................................................................................... 38

FIGURE TWO ....................................................................................... 40

vii

CHAPTER ONE
LITERATURE REVIEW
Singing
Importance of Singing

Many adults, and children as well, take for granted the many venues that include
singing; people sing in church, at sporting events, with the radio, around the campﬁre,
and in the shower (Rutkowski, 1996). Yet, many adults have never found their singing
voice. In spite of all the techniques music teachers use to help children learn to sing,
some leave school unable to use their singing voice properly (Rutkowski & Miller, 2003).
Contrary to popular belief, children are not born with a ‘singing voice’ (Gordon, 2003).
Singing is a skill that can be learned. All children, barring a physical handicap, can be
taught to sing, with the most success when developmentally appropriate instruction
begins in early childhood (Gordon, 2003; Gould, 1969). Because many children cannot
ﬁnd their singing voices on their own, it is up to the practicing music teacher to help
guide them toward appropriate use of their voices (Rutkowski & Miller, 2003).

Singing has always played a central role in the elementary music curriculum and
has been used to teach most concepts in elementary general music (Green, 1989). For
over ﬁfty years, researchers have been studying children’s singing ability (Goetze &
Horii, 1989) and have come to the conclusion that all children can be taught to sing
(Rutkowski, 1996). The National Standards for Arts Education included singing as the
ﬁrst of nine standards to be met when providing a quality music education to students.

According to the standards, students in grades K-4 shall “. . .sing alone and with others, a

varied repertoire of music ...independently, on pitch and in rhythm, with appropriate
timbre, diction, and posture and maintain a steady tempo.” (MENC, 1994).

Although research in singing has spanned much of the 20th century, still there is a
need for continued investigation (Yarbrough, Bowers, & Benson, 1992). Brophy (1997)
states that one ﬁfth of all elementary students do not sing on pitch. If many elementary-
aged children never learn to use their singing voice, this must be seen as a serious
problem for music educators to solve (Yarbrough, Bowers, Benson, & Green, 1991).
Children who experience problems with ﬁnding their singing voice must be properly
diagnosed and treated early. If the problem is not addressed early on, the child runs the
risk of developing unfounded psychological beliefs that he or she cannot sing (Gordon,
2003). Students may begin to associate negative feelings and a bad attitude toward
singing if they do not learn to control their singing voice (Green, 1989), and this negative
attitude may never be reversed (Rutkowski, 1990). Finding singing voice is one of the
ﬁrst steps a child needs to make musically. If he or she does not take this ﬁrst step,
music learning will become more and more difﬁcult. This, in the end, often results in a
poor attitude toward music (Gould, 1969).

Rutkowski (1996) has found that approximately eighteen percent of children in
grade six and below are considered problem singers and cannot sing simple songs. These
children are oﬁen labeled monotones. She goes on to say that these ‘monotones’ cannot
successﬁrlly participate in the most basic singing activities expected by the general
population. As singing can be a social activity, those who cannot use their singing voices
may often feel left out or embarrassed. These children growing toward adulthood are

deprived of one of the most personal forms of expressing themselves.

Singing Voice Development

Until recently, most researchers were overlooking that singing voice use is
different from singing intonation. However, Rutkowski (2004) has found that use of
singing voice is not meaningfully related to pitch discrimination or music aptitude. It is
possible that a child may have highly developed pitch skills and be unable to use his or
her singing voice. Likewise, it is possible that a child will be able to use his or her
singing voice and be unable to match pitch or sing in tune.

Intonation is but one of the many complexities of the singing voice. A child
needs to be able to control his or her singing voice before a teacher can begin to address
intonation problems (Rutkowski, 1990). Singing ‘in-tune’ and a child’s maturing are
linked together. Tuneful singing is a psychomotor process; singing voice is both
kinesthetic and psychological (Levinowitz et al., 1998). A child must ﬁrst differentiate
the motor skills involved in sustaining vocal sounds from those used in speech (Gould,
1969). Gordon (2003) states that the difference between speaking voice and singing
voice is more than the vocal range used when speaking or singing. The difference
between speaking voice and singing voice is the tone quality of the voice due to the way
the sound is produced. When using a speaking voice, the air ﬂow starts in the vocal
chords. While, when using a singing voice the air ﬂow starts in the lungs with support
from the diaphragm and ﬂows through the vocal chords. After a child ﬁnds his or her
singing voice, the child must ﬁnd the motor skills involved in ascending and descending
of melodic contour (Gould, 1969). Melodic contour must ﬁrst be established before a

child can move onto overall key stability (Welch, 1994).

Gordon (1971) states, many times teachers put people into two categories: people
who can sing and people who cannot sing; yet there are many different types of ‘singers’
between. People can be classiﬁed as singers who try to sing in a speaking voice range,
singers who try to sing higher than the normal singing range, singers who sing out-of-
tune and have no sense of melodic contour, singers who sing-out-of-tune and have a
sense of melodic contour, singers who can sing in a group but not on their own, and
singers who can sing alone but not with a group. Gould (1969) collected survey results
from teachers on their perception of the common problems of poor singers. Five
categories were deﬁned: the too low singer, the too high singer, the one-note singer, a
_ combination of the three, and the psychologically inhibited singer.

To gain full use of one’s singing voice, one must employ three separate registers:
the high register, often referred to as head voice, the middle register, and the low register,
often referred to as chest voice. The low register is the easiest to ‘ﬁnd,’ as this register is
what is used for speaking voice, while the high register is the hardest to ‘ﬁnd.’ One can
determine the register in which one is singing by the quality of the sound being produced,
and, to a lesser degree, by the range of pitches being sung (Rutkowski, 2004). Often,
children who sing poorly are trying to sing using a Speaking voice quality in a speaking
voice range (Gordon, 2003). Therefore, many children that are labeled as not being able
to sing are trying desperately to use the three-note range classiﬁed as the speaking voice
range, simply because they have not learned to use the other parts of their singing voices
(Gordon, 1971, 2003).

Gordon (2003) states that there are two types of immature singing voices. With

the ﬁrst, a child tries to sing with a speaking voice quality in a speaking voice range and

with the other, a child sings in a squeaky speaking voice in a range far above the normal
singing voice. The second is much less common. The speaking range is A below middle
C to middle C. The most natural range for the singing voice is from D above middle C
and above, with a break occurring around B ﬂat that carries the singer into what is most
commonly known as head-voice. When children initially ﬁnd their singing voice, their
range will be from D above middle C to A, favoring F sharp and G. Initially, to push
above the break or below middle C will cause tightening of the throat muscles. This
often causes a child to revert to using their speaking voice.

After ﬁnding initial singing range, children often sing out of tune. There are two
types of out-of-tune singers. A child can lack a sense of melodic contour (Gordon, 1971 ,
2003) and a sense of pitch or resting tone. Otherwise, a child can have a sense of melodic
contour but has no sense of pitch or resting tone. The second is harder to ﬁx, for singing
in tune is not a matter of how pitches relate to each other, but rather how each pitch
relates to the resting tone (Gordon, 2003). Singers who sing out-of-tune have not
developed musically their sense of tonality, which results in a poor sense of melodic
contour and pitch (Gordon, 1971, 2003). Welch (1994) agrees that many children will
ﬁnd the melodic contour of a song before coming into key stability. A child’s range will
increase as he or she learns to sing in-tune in the initial singing range (Gordon, 1971,
2003). After children have good use of their singing voice in the initial singing range,
they may proceed to learning songs in the singing range from middle C to high D
(Gordon, 2003). A child needs to feel conﬁdent in his or her conversational singing
range before they can progress farther. If not, the child might regress in singing voice use

instead of making progress (Gould, 1969).

Assessing Singing

Criterion Measures

Starting in the 1970’s, teachers have been held more accountable for what they
teach with each passing year (Levinowitz et al., 1998). In 1994, the National Standards
for Arts Education were established, raising the bar even higher for music educators
across the country (MENC, 1994). These standards provided music teachers clear
expectations for the skills their students should be able to demonstrate at speciﬁc grade
levels. As a result, some teachers found that they were not adequately equipping their
students musically and had to refocus some of their teaching (Byo, 2000; McGuire,
2002). Accountability and the standards created an even greater need for quality
assessment tools in music education (Levinowitz et al., 1998). Valid methods of
measurement need to be developed to measure children’s use of their singing voices
(Geringer, 1983). Many different methods of measuring singing voice use have been
developed and shown to be valid in research settings. Much of this research has focused
on tessitura (Levinowitz et al., 1998), pitch discrimination and pitch matching (Brophy,
1997; Geringer, 1983; Goetze & Horii, 1989; Green, 1989; Jordan-DeCarbo, 1982;
Moore, 1994; Yarbrough, Bowers, & Benson, 1992; Yarbrough, Green, Benson, &
Bowers, 1991), vocal model (Green, 1989; Yarbrough, Bowers, & Benson, 1992;
Yarbrough, Green, Benson, & Bowers, 1991), age (Geringer, 1983; Green, 1989; Moore,
1994; Yarbrough, Green, Benson, & Bowers, 1991) and gender (Jordan-DeCarbo, 1982;
Moore, 1994). Welch (1994) states there are two basic forms of singing assessment:

human and machine-based.

Geringer (1983) used a machine-based measurement when he investigated the
relationship between pitch-matching and pitch-discrimination abilities. Subjects in his
study were (n = 144) four and ﬁve-year-olds from an ethnically diverse preschool and
fourth graders from ﬁve different public schools. After spending three days in their
classrooms to establish rapport with the students, Geringer tested the students
individually on pitch-discrimination and pitch-matching abilities. For the pitch-
discrimination test, three tapes were produced on a Johnson Intonation Trainer in the
vocal range of the children with varying order of twelve tonal pairs. All tapes were
monitored by a Conn Chronatice Stroboscope for pitch accuracy. Each tone was one
second long with .1 seconds between the pair and ten seconds to respond verbally. The
children listened to the recordings and were asked if the second tone heard was the same
or different than the ﬁrst, or if they were uncertain.

For the vocal pitch-matching test, Geringer used a simple unaccompanied three-
measure song with simple words. The song included three pitches and started and ended
on tonic. The song was recorded in three keys by a female singer viewing a strobotuner
to ensure pitch accuracy. Further, a panel of graduate students in music education
selected the most accurate recordings made in each of the three keys while also watching
the strobotuner. After each child listened to the recording, that child was told to sing
back the last pitch he heard after the song was played in each of the three keys. The
responses were tape recorded for analysis using a strobotuner. Geringer found no
signiﬁcant difference between groups for the pitch-discrimination test, while the pitch-
matching test showed a signiﬁcant difference between the groups, with the fourth graders

matching pitch more correctly than the preschoolers (Geringer, 1983).

Green (1989) studied the effect of vocal model on pitch-matching accuracy. The
purpose of his study was to determine the effect of female, male, and child vocal
modeling on pitch-matching abilities of grade school children. Green also used a
machine to measure pitch accuracy.

Audiotapes were made of a female soprano, male tenor, and a nine-year-old child
singing a descending minor third on the syllable ‘la’. Two-hundred eighty-two
elementary students in grades one through six were tested individually three times at
seven day intervals. At each testing all factors were the same except for the vocal model.
Each child was asked to echo the descending minor third after the presented vocal model
at each session. The children’s responses to the vocal model were recorded. A Korg
Auto Chromatic Tuner, Model No. AT-12 was used to evaluate the pitch accuracy of
each response (Green, 1989).

Green (1989) concluded that vocal model had a signiﬁcant effect on pitch-
matching ability. The child model resulted in the most correct responses, followed by the
female model, with the male model resulting in the least amount of correct responses. Of
the six grades tested, children in grades one and six had the highest percentage of
incorrect responses.

Many researchers have investigated the effect of different vocal models.
Yarbrough, Bowers, & Benson (1992) studied vocal models as well, focusing on the
effects of vibrato on pitch-matching accuracy. Tapes were made of a child model, a
female model with vibrato, and the same female model with no vibrato, all singing a

descending minor third. The tapes were analyzed using a digigrarn MIDIMIC and

Macintosh SE computer system for accuracy. Two-hundred kindergarten through third
grade students were tested individually, echoing each model.

Twenty-ﬁve percent of the responses were sent to an independent judge to
evaluate. There was a ninety-seven percent agreement between the experimenter and the
independent judge. The child model resulted in the most correct responses, followed by
the no-vibrato model. Also, certain singers had a high percentage of correct responses
regardless of gender, grade level, or model (Yarbrough, Bowers, & Benson, 1992).

All of the studies discussed so far required a speciﬁcally designed test outside of
the classroom setting to measure vocal accuracy. In an effort to provide information that
is useful to the classroom music teacher, Goetze & Horii (1989) used live performances
to assess students in as similar situation as possible to the ‘normal’ music classroom. The
study focused on accuracy of pitch singing in a group and singing individually.

One-hundred children in kindergarten through third grade participated in this
study. The investigator established rapport with the children before the testing and
presented the melodies by rote that the children would encounter during subsequent
testing. The students were tested in small groups. Each student had their own contact
microphone that recorded them at the same time on separate tracks by using a four-track
tape recorder. The researchers played the responses back into a ‘Visi—Pitch’ machine that
made a graphic of the singing and gave it a number value to represent skill in pitch
matching accuracy. Nine months later, the researchers went back to their data with an
additional judge that reviewed the scores again. Results showed that the children sang
more closely to the pitch when singing individually than in a group. Third graders sang

more accurately most often. Girls sang more accurately than boys, especially in group

singing (Goetze &Horii, 1989). Although their testing was in small groups, it still
required a considerable amount of equipment, making it impractical in most classroom
settings.

Yarbrough, Bowers, Benson, & Green (1991) studied vocal models and response
modes. One-hundred sixty-three children were chosen by a music teacher for their
inability to match pitch to a previously recorded male and female model singing a
descending minor third. All of those children then received instruction in Kodaly hand
signals and solfege syllables as they performed. Two weeks after the initial testing,
students were tested again with the two models, but this time they used the Kodaly hand
signals or solfege syllables. Judges rated the performances. There was no difference
between the Kodaly response modes. However, children were able to match the female
model better than the male (Yarbrough, Bowers, Benson, &Green, 1991).

There are many things that can be gleaned from the above studies. Research has
focused on many different aspects of singing, mostly focusing on pitch or pattern
matching outside of a musical context. Much time, energy, resources, and expensive,
ﬁne-tuned equipment have been essential to the success of these studies. Other than with
the study by Goetze & Horii (1989), no attempt has been made for testing to take place
under the ‘normal’ conditions, encountered in a typical general music classroom. Gould
(1969) states that there is little reason to carry out research if it does not reach and beneﬁt
the classroom teacher. However, none of these studies include a measure of singing

voice that would be practical in a ‘real world’ teaching situation.

10

Practical Assessment

Assessment of vocal pitch accuracy has been approached in a variety of ways.
Researchers have used the precise measurements of computers and employed highly
skilled judges to use their rating scales. Children have been asked to sing both familiar
and unfamiliar songs, as well as patterns and individual pitches. They have been tested
privately aswell as in a group (Brophy, 1997). Yet, there still exists a need for practical
assessment tools for the elementary general music teacher to use in measuring singing
voice in their classroom. Goetze, Cooper, &Brown (1989) report that stroboscopic and
oscilloscopic devices provide the most accurate and objective measure of pitch, for they
give a precise evaluation of singing accuracy that teachers can not make on their own in
the classroom. However, elementary music teachers need to assess many students. This
is difﬁcult, because they see their students once or maybe twice a week in a group and do
not have the time to assess students privately or even in small groups (Brophy, 1997).
Also, those devises measure absolute pitch rather than pitch in the context of a tonality,
which are slightly different. Using technology to analyze pitch is time consuming and
impossible with a large group of students (Rutkowski, 1996); it is also expensive and
requires computer software that is not readily available (Brophy, 1997). 1

Formal singing assessments incorporating various measurement devices are
different from what a child typically experiences in the general music classroom, making
them less valid. This can have a negative effect on a child’s ability to understand the
singing task and demonstrate singing proﬁciency (Brophy, 1997). Elliot (1995) states

that assessment should be embedded in the process of learning. Although researchers

11

have studied the factors effecting singing accuracy, there is little research on how to
judge vocal pitch accuracy in the context of a normal music classroom (Brophy, 1997).

If practicing elementary music teachers are to meet the singing benchmark
outlined in the National Standards, there must be practical and accurate assessments that
they can use to measure the singing achievement of their students. To this end, Goetze,
Cooper, & Brown (1989) suggest a study of the relationship between a subjective rating
scale and an objective acoustic analysis would be useful.

Children enjoy games and sing alone with more case and conﬁdence in a playful
environment (Brophy, 1997). Singing games are already incorporated into most
elementary general music classrooms (Brophy, 1997). Practicing elementary music
teachers might be able to assess students vocal pitch accuracy by using singing games
without creating an artiﬁcial environment, as children sing alone during many traditional
singing games. Children might not know that they are being assessed while playing the
game, so they would not get nervous; this results in a truer evaluation of the child’s vocal
achievement (Gagne, 1997).

Brophy (1997) conducted the only study investigating whether singing games can
be used when measuring vocal pitch accuracy. Brophy found that using children’s
singing games to measure pitch accuracy was effective. He also found that older students
were more consistent in singing accurate pitches than younger students when using
singing games as assessment tools. In Brophy’s study, he measured the accuracy of
speciﬁc pitches within the singing games. Two songs with the same target pitches were
used in the assessment to measure validity of the singing games as a measure of vocal

pitch accuracy in the general music classroom. The rating consisted of a + or -. The

12

child either sang the correct target pitches at the correct pitch level, or they did not.
Teachers need to know the level of singing voice development if they are to meet the
instructional needs of each child. He concluded that singing games can be considered to
be a valid tool in measuring of singing accuracy. However, whether children can use
their singing voices is not a ‘yes’ or ‘no’ question.

All of the measures that are available, regardless of whether they are practical for
the music teacher, have either focused narrowly on intonation or have combined the
aspects of singing in-tune and singing voice achievement into one measurement.
However, a rating scale should measure one thing at a time if it is valid (Gordon, 1971).
To date there is only one measurement tool that focuses on measuring the development of
a child’s singing voice, as opposed to measuring pitch accuracy or intonation. That is
Rutkowski’s Singing Voice Development Measure (SVDM). She developed this
measure based on the belief that use of singing voice is a separate behavior from the
ability to sing in-tune (Rutkowski, 1996). One of the reasons Rutkowski developed
SVDM was to provide a more consistent structure when identifying various stages of a
child’s singing voice (Rutkowski, 1990). This is important to practicing music teachers
and researchers alike, because this information can be used by elementary music teachers
to provide more appropriate singing voice instruction. Rutkowski talked with many child
voice experts before designing her ﬁrst draft of SVDM (Rutkowski, 1990). SVDM has
been shown to be a valid measure of singing voice achievement (Levinowitz et al., 1998;
Rutkowski, 1990; Rutkowski, 1996; Rutkowski & Miller, 2003).

Rutkowski’s measure initially had ﬁve levels to describe a child’s singing voice,

but she expanded it after receiving feedback from raters used in her pilot study, during

13

which they stated that they observed that children often bordered between two of the
levels on the rating scale. She chose to keep the continuum for the rating scale under ﬁve
for ease of rating. The following is the scale as it is now used: 1- Presinger, 1.5 —
Inconsistent Speaking Range Singer, 2 — Speaking Range Singer, 2.5 — Inconsistent
Limited Range Singer, 3 — Limited Range Singer, 3.5 — Inconsistent Initial Range Singer,
4 — Initial Range Singer, 4.5 — Inconsistent Singer, 5 — Singer. Rutkowski also developed
a set of vocal patterns to be echoed by children that include all of the ranges of a child’s
singing voice, so that one can get a complete picture of a child’s singing voice
development using her scale (Rutkowski, 1990).

Even with the success of SVDM in measuring singing voice achievement,
Rutkowski admits that testing individually takes a great deal of time and is tedious
(Rutkowski, 1996). Welch (1994) believes that children are often misdiagnosed as poor
singers in research studies because of the formal nature of the assessment procedure as
opposed to the natural environment the child is accustomed to in the general music
classroom. Gould (1969) suggests that there is a loss of communication somewhere
between the ﬁndings of research and the classroom. Consequently, there is still a need
for a tool for measuring singing voice development that music teachers can easily use in
their general music classroom.

Brophy was successful in using singing games as a venue through which to
measure intonation while in the music classroom (Brophy, 1997). Rutkowski developed
a reliable assessment for measuring singing voice achievement, but it is difﬁcult to use in

a classroom setting (Rutkowski, 1990). Could one successfully use the rating scale

14

developed for SVDM in assessing singing voice achievement in the context of singing

games, during which children are in a natural and uninhibited environment of play?

Young Children’s Growth and Development

Play

Since the beginning of man, play has been seen as a dominant behavioral aspect
of our species, speciﬁcally if one looks at our children (Brophy, 1992). Children express
themselves through play by exploring the world around them. Through play, they
rehearse things that have already taken place, as well as things to come, practicing skills
and reﬁning them. Play helps them to learn and assimilate important knowledge. In
short, they use play to make sense of the adult world around them (Van Der Linde, 1999).
Piaget (1962) describes children’s quest for cognitive development as ‘trial-and—error’
behavior. Children work out this ‘trial-and-error’ behavior through play. Neuroscientists
have discovered that play, once seen as frivolous and unimportant, is actually as
important as other basic drives, such as sleep, rest, and food (Frost, 1998).

Researchers have been studying play for many decades (Berger & Cooper, 2003).
Frost (1998) reports that play is important for children’s brain growth and development.
Between the ages of three and ten, a child’s brain contains two times more synapses than
that of an adult. The play process during childhood reﬁnes the uses of the synapses and a
child’s intellectual ability for life. Brain development is a ‘use it or lose it’ process, says
Frost. Synapses that are not activated in a young child through their experiences will
cease to exist. In earlychildhood, a child’s brain is highly inﬂuenced by environmental

stimulation, and ‘playﬁil activity’ makes a positive difference in how the brain develops

15

(Frost, 1998). Because neural synapses in the brain and a child’s music aptitude are
developmental, they must be nurtured in early childhood. Otherwise, the result of neglect
will be a lifetime of less than optimal potential (Taggart, 2003). Frost (1998)
recommends that one read, sing, and play with children everyday in order to help their
brains develop properly.

Children are not adults, and it is foolish for educators to try and teach children as
if they were. Children have their own unique way of learning, and educators need to
know how best to address their learning processes (Taggart, 2003). When children are
relaxed and engaged, they are able to learn in the most natural way (Yurko, 1992).
Children become entirely absorbed during activities they consider ‘fun,’ but if the
enjoyment element is taken away, the play they were previously absorbed in stops
(Addison, 1991). Students who are enjoying themselves forget that the focus in the
beginning was to learn. Personal experience and exploration through play should be the
basis for learning, instead of adult imposed traditional perceptions (Berger & Cooper,
2003).

The National Association for the Education of Young Children promotes play and
informal experiences for children in early childhood (Bredekamp & Copple, 1997). Play
is a child’s attempt to master their environment. When children play, they are
intrinsically pulled into the joyful actions, for it is the process that is the focus, not the
product (Berger & Cooper, 2003). Young children learn by doing, through activity,
through play (Brophy, 1992). Children can be seen acting out the dramas of life while
they play. A little girl might pretend she is cooking a meal for a host of stuffed animals.

A little boy might ﬁx a table using his plastic power tools. Addison (1991) states that

16

drama and stories are merely miniature enactments of real life, and those who participate,
both the actors and the audience, all knowingly share in the illusion of reality.

Educators can glean insights into how a child is developing cognitively, socially,
physically, and emotionally, just by watching him play (Taggart, 2003). Piaget agrees
that play is an indicator of children’s cognitive development and is a way for us to
‘know’ where they are developmentally (Piaget, 1962).

So what exactly is play? The word ‘play’ has many different meanings. What
one person considers play, another may not. Addison (1991) lists eight different
“Properties of Play” to help distinguish play from other behaviors.

1. People work at play, some harder than others.

2. People put immense effort into play, play is usually

treated seriously.

3. People like to play. It is what makes life worth living.

4. Play is self-contained. It is intrinsically interesting.

5. Play is voluntary.

6. Play makes people concentrate; it absorbs people when they play.

7. Play is in the control of the players involved.

8. Play is symbolic; it is not real life.
Frost (1998) supports these properties of play when he says that, play is more than a
ﬁivolous activity; there is something deeper in play than what ﬁrst meets the eye. Playful
activities in which children engage serve to provide them with more than just enjoyment,
but also provide them practice for acquiring both physical and social skills that they will

use later in life (Brophy, 1992). Children rarely sit still, for play is their naturally active

17

mode of living. Teachers should not try and stop these playful activities, but rather, give
children the freedom to play; play is the most natural way for them to learn (Addison,
1991). Play should serve as a core, rather than a fringe in early education. Educators
need to set up environments that enhance play and then guide children to learning
through the experience themselves (Taggart, 2003).

One of the ﬁrst educators to recognize play as a valuable teaching tool was
Friedrich F roebel (Brophy, 1992). He considered play to be central in the development
of children (Berger & Cooper, 2003). Froebel’s learning philosophy was that children
should be active learners. On that principle, he founded the ﬁrst kindergarten, meaning
children’s garden, in 1837 after tutoring children in a garden and watching them learn
through action. Before Froebel’s kindergarten, there was no school for children under
seven years of age. People of that time believed that children could not develop cognitive
or emotional skills so early in life. Froebel believed that children were capable of these
things and should follow their own interests and explore them through what he called
‘self-activity.’ The role of the teacher was to be a guide in ‘self—activity’ (“Friedrich
Froebel: Founder, ” 2000). The following were Froebel’s kindergarten goals: “physical
activity, development of sensory awareness and physical dexterity, creative expression,
exploration of ideas and concepts, the pleasure of singing, the experience of living among

others, and satisfaction of the soul” (p. 63). All of these occur in and through play.

Play verses Work

Work is a purposeful, directed activity aimed at achieving a predetermined goal

(Brophy, 1992). Most people would be able to identify the difference between work and

18

play if they saw it, but when describing it, have difﬁculties distinguishing play from non-
play (Van Der Linde, 1999). Sometimes the line between work and play is somewhat
blurred. What some consider work, others consider play, and vise versa. Play cannot be
distinguished by where it takes place or for whom it is done (Kratus, 1997). Children do
not distinguish clearly between work and play. They are learning life’s necessities
through play (Addison, 1991). Often the words ‘work’ and ‘play’ can be used
interchangeably, making what is work and what is play conﬁising, for these words seem
to suggest such polar extremes; yet, they can be used to refer to the same phenomenon
(Kratus, 1997). When a child must concentrate ﬁtlly on the activity at hand, the concept,
learning, understanding, and retention naturally fall into place in the child’s mind,
making the distinction a non-issue (Hotchkiss & Athey, 1978).

Kratus (1997) identiﬁes some similarities and differences between work and play.
Both work and play are intentional and purposeful activities. Work is an activity to
achieve a particular outcome and has real world ramiﬁcations. Play is an activity that
sparks imagination in which reality is somewhat suspended. One will choose to play or
work for entirely different reasons. People may play for the experience or work for the
sense of accomplishment. People have many biases when it comes to work and play.
Many times work is viewed as something bad, yet productive, whereas play is viewed as
something good and enjoyable, yet unproductive (Kratus, 1997). When the qualities of
work and play are meshed, one often ﬁnds activities that are inexpressibly satisfying
(Csikszentrnihalyi, 2000).

When one looks at the things that give people the most satisfaction and worth, he

ﬁnds aspects of both work and play. So, it seems logical that music education can be

19

most effective and beneﬁcial when music educators combine aspects of work and play.
Teachers should strive to combine the elements of work and play in the classroom to
achieve a higher level of musical grth (Kratus, 1997). Music instruction should
include interesting objects, ideas, and activities that possess the qualities of both work
and play. When teachers have achieved this, they will no longer have to try to get

children to learn or to practice; children will want to naturally (Addison, 1991).

Music and Play

Play is the true origin of all of the arts that we enjoy today (Brophy, 1992). Plato
stated in 300 BC that the most powerful educational aid was music (Van Der Linde,
1999). Music is a form of play (Addison, 1991). It is natural for a child to sing and play;
it is vital to their development. Young children ﬁnd music and play to be inseparable
activities. Berger & Cooper (2003) state that it is natural for children to respond to music
while playing. When young children play, they often tend to sing or hum spontaneously
(Van Der Linde, 1999). Movement can also be seen accompanied with vocal sounds or
music for children, as children have not yet acquired the ability to sit still when
meaningful music is being experienced (Addison, 1991).

Children should participate in musical play, because it facilitates their musical
grth (Berger & Cooper, 2003). Children learn music in much the same way as they
learn language. As children are bathed in a verbally-rich environment in which they
learn to master language, so should they be bathed in a music-rich environment, with
plenty of musical ‘toys’ to play with and manipulate. It is through this musical play

environment that children begin to impose order to and gain understanding of the sounds

20

around them (Taggart, 2003). Musical play allows children the freedom to explore,
improvise, and create (Berger & Cooper, 2003). Frost (1998) states that long—term
changes in neural structures effecting thinking and reasoning abilities can be linked to
play, art, and music.

Many parents do not sing and move with their children. Music educators need to
step in and provide playful and child—centered music environments in addition to formal
music training. Music educators need to allow children time to explore and play in their
own way, while encouraging them in their play. A child’s music development can be
hindered by adults expecting traditionally ‘correct’ music responses too early (Taggart,
2003). Berger & Cooper (2003) agree; a child’s version of musical play is often not
what adults would consider music making. Musical play is enhanced when adults refrain
from correcting children in their method of free musical play. Adults can enhance
musical play further by encouraging all music exploration. This encouragement can be
both verbal and non-verbal. Adults can encourage a child’s musical response through
positive eye contact, echoing musical utterance, and extended musical dialogue that is
child-initiated and child-directed (Taggart, 2003).

Feierabend has based much of his curriculum at The National Center for Music
and Movement in the Early Years at the Hartt School of the University of Hartford on
musical play. The music and movement classes consist of ﬁfteen minutes of ‘free
musical play,’ followed by twenty to thirty minutes of circle time. The aim for the
children is development of comfortable and accurate singing skills, comfortable and
accurate moving skills with beat in duple and triple meters, and expressive sensitivity.

All of this is achieved through some form of musical play (Feierabend, 1998).

21

Berger & Cooper (2003) identiﬁed Three Themes of musical play. The ﬁrst
theme was Unﬁnished Play. Unﬁnished Play is when free musical play is interrupted, yet
the children express the desire to continue playing. The second theme was Extinguishing
Play. Extinguishing Play is when adults or another child did something that stopped the
child’s free musical play completely. The ﬁnal theme that they identiﬁed was Enhancing
Play. Enhancing Play is when adults encouraged children’s free musical play and further
musical growth was seen. If a child is left alone to organize his own free musical play, it
helps him progress from the free musical play to higher discerning musical skills. Berger
& Cooper also found that, given the opportunity, many children will engage in free
musical play even if there are other, more structured musical activities going on around
them.

While free musical play is said to enhance musical growth, Taggart (2003) warns,
it is not enough to provide children with music centers ﬁlled with percussion instruments
if the children do not have the tonal and rhythmic vocabulary to apply to their play,
making it a meaningful experience. Teachers need to know where students are
developmentally musically to best serve them.

Young people want to learn, to improve in skill, and to grow. They become
absorbed in an activity and will put a lot of effort into the activity if the play is attractive
enough (Addison, 1991). Musicians talk about how they ‘play’ an instrument, but it is
many times the hard ‘work’ they put in on learning an instrument that allows them to
‘play’ well (Kratus, 1997). Brophy (1992) states that young children want to gain power

over their world, and, in doing so, they naturally delight in ﬁnding sounds and mastering

22

them. Mastery of sound in music can be worked toward through singing activities and

games, all of which can be a part of musical play.

Games

Hotchkiss & Athey (1978) describe a game as “. . .any structured activity in which
there is an element of suspense and challenge and in which the participants enjoy
themselves” (p. 48). Math and reading have been using games for many years to teach
concepts (Hotchkiss & Athey). In childhood, children begin to organize and systemize
the objects around them (Piaget, 1962). The ordering of games with rules begins at
approximately the same time. Games are a child’s natural way of thinking. As a child
works to complete the steps needed to reach the end of a game, he is developing both
mentally and physically in ways that he is not often aware (Hotchkiss &Athey, 1978).

Games fall in the hazy area between work and play. In a classroom, a child often
does not distinguish that they are working to learn a new concept when engrossed in a
game; they view the activity as play. When comparing games and play, it is easy for
adults to see the difference between the rigidity of games and the ﬂexibility of
unstructured play (Prim, 1995). Games give a perimeter in which to work and learn new
concepts. Play is more spontaneous and creative (Dalby, 1992). A child who ‘plays’ too
much in music will not achieve anything meaningful (Kratus, 1997). This is when the
teacher should step in and give some boundaries and structure to stay within and a
destination to work toward, for the children will still view it as play. Games that offer
variety and excitement are one of the most effective teaching methods for deve10ping

musicianship (Dalby, 1992). Frost (1998) agrees that teachers need to initiate children to

23

play, giving the play structure and purpose with games. Activities such as singing,
dancing, and acting are considered games, because they are organized play (Prim, 1995).

Prim (1995) states that children become interested in games that involve music
and movement as early as ages three or four. Children enjoy playing games that they
associate with music (Van Der Linde, 1999). In music class, games are engaging
activities that help children build musical and social skills. They can be used to help
improve listening skills and help children learn to sing melodies (Howle, 1997). Games
equip children with skills that they will need later in life and enable them to practice
those skills in a non-threatening environment (Frost, 1998). Games can uncover concepts
that students do not understand. In a game, students are less concerned about making a
mistake, and they can help the teacher diagnose learning strengths and weakness
(Hotchkiss & Athey, 1978). In games, age, gender, or class are not barriers to
participation; games break down these barriers (Howle, 1997). Hotchkiss & Athey
(1978) agree that games can dissolve barriers, allowing children to be themselves. When
games are played in a circle, no one predominates (Howle, 1997). Games provide non-
risk situations in which children can ‘try themselves out.’ Games enable children, as a
part of their natural development, to ﬁnd out what they are good at and to improve those
skills in a non-threatening environment. (Addison, 1991). Children who at ﬁrst are too
shy to participate will soon be unable to resist the enjoyment and excitement of a game
and join in enthusiastically (Howle, 1997). Gange (1997) agrees that children, who are
often reluctant to participate in rote singing, will gladly join into the activity when

playing a game.

24

Many well-known methods of teaching music have included games in their
philosophy of teaching (Hotchkiss & Athey, 1978). Montessori schools incorporate
games in their teaching (Beaven, 2000). Teachers that use the Orff method in their
classrooms use games to build rhythmic unity and prompt body movement (Shamrock,
1997). Kodaly also supported the use of games (“About Kodaly,” 2004). Prim (1995)
identiﬁes six elements that can be taught through musical games:

1. The difference between speaking and singing.
2. Songs with three, four, ﬁve and more different notes.
3. Songs with larger or lesser range.
4. Songs with binary or ternary meter.
5. Complex and simple rhythms.
6. Song forms such as A, AB, ABA...
Obviously, this is not an exhaustive list. Most musical concepts can be approached

through games.

Singing Games

“Because play is satisfying in itself and seems to have no consequences, the
learner is freed to experiment and explore without fear. Young children are assumed not
to identify the musical activity being engaged in as anything but a game, and are
therefore psychologically released to participatein relatively uninhibited play for which
the trained specialist has one goal and the children another. Singing games are
particularly useful for ear-training activities, solo singing, and beginning vocal

improvisations...” (Brophy, 1992, p. 12). Mitchell (1991) agrees that songs, games, and

25

activities are valuable practice for developing accurate singing. There are many beneﬁts
to singing games, one of which is that children love to play singing games and are upset
if they do not get a chance to have a turn (Gagne, 1997).

Mitchell (1991) states that, when games have a purpose, children can improve
their singing skills in a positive environment. Singing games have helped countless
children sing in tune. They help the non-singer to become singers and the singers to
become more accurate in their singing abilities. Gagne (1997) agrees that games help
non-singers perfect their sense of pitch and goes further to remind us that it helps their
cognitive development as well.

Feierabend’s (2000) research on young children’s musical development
prompted him to develop a curriculum based on singing games and activities. His quest
began with a strong belief that adults should be able to dance, sing, and express
themselves through music. He believes that many do not, due to insufﬁcient musical
experiences during childhood, which is when music activity should begin and be
nourished. Imbedding music in people’s lives should be a natural cycle, according to
Feierabend. Children should become ﬂuent in music so that they can grow up and, in
turn, teach their children music. Historically, families passed on rich musical activities to
the generations that followed in this natural cycle of musical learning. Feierabend
explains this decline of family musical heritage as a shift in community. No longer does
one see large families living within a mile of each other. Families are smaller and spread
across the globe, succumbing to the ‘ear-candy’ of the current culture. Feierabend’s
approach to music curriculum is to recover folk songs and games of old and teach them in

a playful environment (Feierabend, 1996).

26

The practicing music teacher should have a repertoire of purposeful music games
that promote both group and solo singing in the classroom to enable children to feel safe
when singing alone (Rutkowski, 1996). Play and its role in child development has
received much attention and has been the focus of research (Brophy, 1992). Yet, there
has been little research done speciﬁcally on the use of music games in the general music
classroom (Hotchkiss & Athey, 1978). Prim (1995) remarks that “studies concerning
singing games appears to have stopped...” (p. 149). However, could the strength of
using singing games in the music classroom and the need for practical singing voice

measures be combined?

Purpose
With the intent of providing a valid assessment tool for practicing elementary music
teachers, the purpose of this research is to investigate the measurement of singing voice

development through singing games.

Problem
The problem of this study is to determine the validity of using children’s singing games
as a context through which to measure singing voice development in young children,

using the rating scale of Rutkowski’s Singing Voice Development Measure.

27

CHAPTER TWO
RELATED RESEARCH
Singing Voice Development Measure
When Rutkowski began developing her Singing Voice Development Measure

(SVDM), she was concerned with the many names and labels that have been given to
children that cannot gain control of their singing voice: . monotones, inaccurate singers,
backward singers, non-singer, problem singer, and partial singer. These labels often
point toward these children being hopeless cases, not possessing the ability to learn or to
be taught how to gain use of their singing voices. So, when Rutkowski labeled her rating
scale levels, she was careful to use labels that suggested that the process a child goes
through, in the quest of becoming a full ﬂedged ‘singer,’ is developmental. Rutkowski
spoke with several child voice specialists after reviewing past scales that had attempted to
rate singing voice. These scales did not separate the measurement of intonation from the
measurement of singing voice development. Rutkowski (1990) found that these are
separate constructs and therefore should be measured in isolation. Rutkowski articulated
the following ﬁve levels of singing voice development and used them in her pilot study
(Rutkowski, 1990).

1 Children who use only speaking-voice inﬂection but do

not sustain tones
2 Children who exhibit use of melodic contour and sustained tones, but
use speaking range or a very high range
3 Children who use a very limited singing range, usually D3 to F#3.

4 Children who use initial singing range, usually D3 to A3

28

5 Children who are able to sing over the register lift, F b3 and above

Rutkowski conducted a pilot study to determine if the behaviors in her rating
scale were valid. She used a song entitled “Bakerman,” with which the children were
familiar, because she believed that one should use songs familiar to children when
measuring their singing voices. If the songs were unfamiliar, children may become
insecure and not sing to their full potential, thus resulting in an inaccurate measurement
of their singing voice development (Rutkowski, 1990).

The researcher tested students individually singing the familiar minor song. They
were not given a tonality, starting pitch, or tempo before they were recorded. Rutkowski
chose to test this way, because she believed it would result in a truer picture of the child’s
singing voice. Also, she believed that children who are secure with their singing voices
will most often sing a familiar song in the same key in which they learned it. A few days
later, the children were recorded a second time (Rutkowski, 1990).

Four raters who were familiar with young children’s singing voices participated in
the pilot study. Practice examples were provided for the raters at the beginning of the
recordings to familiarize them with the rating scale. Interrater reliability ranged from
.836 to .963. Rutkowski included a ‘ready-sing’ component before a few of the children
to see how that affected the raters. Using the ‘ready-sing’ protocol, the reliability of the
raters was .904. The intraj udge reliability was .918. The raters commented that the
‘ready-sing’ component helped them assess the children. Also, they admitted an
inclination to listen for intonation rather than just singing voice (Rutkowski, 1990).

Rutkowski then proceeded to the main study. The main study was similar to the

pilot study but used a larger and more heterogeneous subject group. However,

29

Rutkowski added tonal patterns to be echoed by the investigator and ‘ready-sing’ in the
same mode and meter on the descending tonic chord of the song as a preparatory
sequence. Patterns were ﬁrst played on the tone bells, then sung by the investigator, then
echoed by the child being tested (Rutkowski, 1990).

Prior to testing, Rutkowski visited the classroom to become acquainted with the
children. She taught them “Bakerrnan” and reviewed the tonal patterns to be used with
them as well. The children’s music teacher reviewed the song with them each week for a
month prior to the testing. The tonal patterns were reviewed with the children a second
time the day they were to be tested (Rutkowski, 1990).

Two of the raters with the highest interrater reliability in the pilot study, ranging
from .836 to .963, were chosen to rate the children in the main study. Rutkowski
determined that SVDM was a reliable measure of the use of singing voice. Following the
study, Rutkowski interviewed the raters to get their feedback concerning SVDM. They
stated that, after they became more comfortable with the rating scale, they much
preferred rating the tonal patterns over the song. Teaching the song takes more time than
echoing patterns. Also, performing a song involves other things besides the singing
voice, such as memorization of the text. As a result, the child may focus on one of the
other components required to sing the song and not demonstrate his or her true level of

singing voice development (Rutkowski, 1990).

Other studies using S VDM
In 1996, Rutkowski conducted another study using SVDM in an attempt to

identify better ways to teach singing voice and develop music aptitude. The purpose of

30

her study was to compare the effect of large-group versa individual/small group singing
activities on singing voice development, to monitor singing voice development over a
time period to see change, and to monitor change in developmental music aptitude over
time. She measured singing voice development using SVDM, as it had been previously
proven to be a valid measure of young children’s use of their singing voice (Rutkowski,
1996).

The subjects were students in intact kindergarten classes that were randomly
assigned to control or treatments groups. All children received the same songs, activities,
and games incorporating singing and movement. In addition, the treatment group had
small-group and individual participation in singing activities. Primary Measures of
Music Aptitude (PMMA) and SVDM were given at the beginning and end of the year,
with PMMA given mid-year as well (Rutkowski, 1996).

The agreement between raters was .90 and .99 for SVDM. Both treatment groups
had higher tonal aptitude at the end of the treatment, but there was no signiﬁcant
difference between groups on tonal aptitude. However, there was a signiﬁcant difference
between the groups on the posttest for the SVDM, showing that the small group and
individual instruction had made a positive difference in the children’s use of singing
voice. Rutkowski also found a weak relationship between singing voice achievement and
tonal aptitude. This points to an even greater need to help children ﬁnd their singing
voice. Some above-average musical aptitude students may be considered by music
teachers and others to be tonally weak, just because they do not know how to use their
singing voices. The raters in Rutkowski’s 1996 study stated that many children were

inconsistent in use of their singing voice. Children also hovered between two stages of

31

SVDM. As a result of their feedback, Rutkowski expanded her original ﬁve-level scale
to include children who do not consistently perform at one level. Instead of adding more
numbers to the scale, she added halves to make it easier for the practicing music teacher
to internalize (Rutkowski, 1996). Her revised SVDM is as follows:

1.0 “Pre-singer” does not sing, but chants the song text.

1.5 “Inconsistent Speaking-Range Singer” sometimes chants, sometimes
sustains tones and exhibits some sensitivity to pitch but remains in the
speaking voice range (usually A2 to C3).

2.0 “Speaking-Range Singer” sustains tones and exhibits some sensitivity

to pitch but remains in the speaking-voice range (usually A2 to C3).

2.5 “Inconsistent Limited-Range Singer” wavers between speaking and
singing voice and uses a limited range when in singing voice
(usually up to F3).

3.0 “Limited-Range Singer” exhibits use of limited singing range
(usually D3 to F3).

3.5 “Inconsistent Initial Range Singer” sometimes only exhibits use of
limited singing range, but other times exhibits use of initial singing
range (usually D3 to A3).

4.0 “Initial Range Singer” exhibits use of initial singing range
(usually D3 to A3).

4.5 “Inconsistent Singer” sometimes only exhibits use of initial singing
range, but other times exhibits use of extended singing range (sings

beyond the register lift: Bb3 and above).

32

5.0 “Singer” exhibits use of extended singing range (sings beyond the
register lift: Bb3 and above).

Rutkowski conducted a longitudinal study of elementary school children’s singing
voices with teacher/ researcher Miller using SVDM (Rutkowski & Miller, 2003).
Twenty-ﬁve children participated in this study throughout their ﬁrst, third, and ﬁfth grade
years. SVDM was administered at the end of ﬁrst, third, and ﬁfth grades. In this study,
the tonal patterns used were sung on the syllable ‘bum’ as well as with text. The
responses of the children were tape-recorded, and the same raters rated the responses
using SVDM after each testing in this longitudinal study. The rater agreement ranged
from .738 to .956. Statistical analysis determined if gains in singing voice were achieved
in each grade. A signiﬁcant difference was found; children’s singing voice development

increased with age (Rutkowski & Miller, 2003).

S VDM tested in a practical environment

SVDM has proven to be a valid measure of singing voice achievement. Yet,
testing is often time consuming and tedious (Rutkowski, 1996). Because measurement
should be a key component in every music program, a practical assessment tool needs to
be available so that a general music teacher can use it on a regular basis in the general
music classroom (Levinowitz et al., 1998). SVDM was created to provide a consistent
means of identifying and measuring the stages of children’s singing voice development.
Seeing worth in the stages of SVDM, Levinowitz et a1. wanted to see if music teachers
could use SVDM in the music classroom, using solo songs instead of tonal patterns, and

across a wide variety of ages and backgrounds (Levinowitz et al., 1998).

33

One-hundred seventy-ﬁve students from ﬁve elementary schools participated in
this study. One month prior to testing, the students were introduced to the two songs by
rote that were to be used for the testing. The two songs were a minor song entitled “In
the Sea” and a major song entitled “Row, Row, Row Your Boat.” These songs were
chosen because of their familiarity across a wide range of ages; also, both songs covered
an octave range from D to D, and one was short while the other was longer to provide
children the opportunity to audiate at a deeper level. The children were taped
individually during the music class in front of classmates by his or her practicing music
teacher. The tonality was established by the music teacher prior to singing by playing I-
V-I chords on the guitar or piano. Then the teacher sang ‘ready-sing’ on the starting
pitch of the song (Levinowitz et al., 1998).

The performances were rated using SVDM by the six co-investigators, all of
whom were practicing music teachers. They found that SVDM is a valid measure for
kindergarten through ﬁfth grade children. The co-researchers of this study recommend
the use of SVDM in assessing singing voice achievement, speciﬁcally when students sing
a short, familiar, major song, based on the interrater reliability (Levinowitz et al., 1998).

Although Levinowitz et al. used SVDM in the classroom setting, the children
were taken out of their natural environment of play and learning. When children got up
to sing, they knew they were being tested on how well they sung in front of their peers.
This may have affected their performances, thus not giving the music teacher a true
assessment of the child’s singing voice development. Music teachers must know their
students’ levels of singing voice development so that they can help them achieve even

higher levels of singing voice use. When children abandon themselves in a singing game,

34

they are not concerned with anything but enjoyment. As a result, their singing maybe
more reﬂective of their singing voice use in natural settings. General music teachers
need a reliable and practical tool to assess singing voice development that can be

successfully used in their classroom. This study is contributing toward that goal.

Validity

In order for the results of research to be considered valid, the criterion test must
be accepted as meaningful and important (Cronbach, 1970). In Rutkowski’s studies, she
based her test’s validity on the judges reliability. The reliability ranged from .836 to .963
(Rutkowski, 1990), .77 to .99 (Rutkowski, 1996), and .738 to .956 (Rutkowski &Miller,
2003). She also worked toward subjective validity by talking with child voice experts
when developing her scale to be certain that her levels accurately reﬂected the stages of
vocal development. A test that is valid for one situation may not be valid for another.
Levenowitz et al. (1998) determined that Rutkowski’s measure was valid for use in their
settings as well.

A correlation coefﬁcient, the relationship between two variables, can be used to
answer the question: ‘Do these two tests measure the same thing?’ If one looks at the
relationship between sets of data from two different measures that were designed to
measure the same thing and they have a strong relationship, the researcher can have
greater faith in the validity of the measures. In establishing concurrent validity, the
scores from a test that has already been proven useful and valid, but for one reason or
another it is considered to be too cumbersome, are correlated with the scores from

another less cumbersome test whose validity is not yet established. The subjects are

35

given both measures in close proximity. In this study, the concurrent validity of using
singing games as a context for measuring singing voice development will be established
by using Rutkowski’s SVDM, a measure already shown to be valid by multiple

researchers.

36

CHAPTER THREE
METHOD

Subjects

Four classes of third grade students (n=78), 37 girls and 41 boys, at a West
Michigan elementary school participated in this study. They are of diverse socio-
economic backgrounds as well as ethnic origins. Each class meets for music class twice a
week for half an hour each time with a trained music specialist, who is also the
researcher. The students are familiar with various singing games and are accustomed to

singing by themselves. Most of the students have a positive attitude toward singing.

Design and Procedures
The study was designed as follows. Students’ vocal performances were rated
using Rutkowski’s Singing Voice Development Measure (SVDM). This portion of the
investigation was modeled after Rutkowski’s (1996) study. During two of the children’s
music classes, students learned and practiced the patterns used in SVDM. The music
specialist taught the patterns according to Rutkowski’s Procedures for Test
Administrator. The procedure follows.
SVDM: Procedures for Test Administrator
1. Play the ﬁrst pattern (measure) on the tone bells or piano
(example: “see the bird”).
2. Sing the ﬁrst pattern for the children with the text or neutral syllable, “bum”.

Do not use any accompaniment.

3. Have the children echo. Again, do not use any accompaniment.
4. Repeat steps 1-3 with each pattern (1 measure equals one pattern).
5. Do not pause in between any of the above steps.

6. For half of the group of children, perform steps 1-5 on text ﬁrst then on the

neutral syllable “burn”; for the other half perform steps 1-5 on the neutral

37

syllable “bum” ﬁrst then on text.

Following in Figure 1 are the tonal patterns used for Rutkowski’s SVDM.

Figure l — Rutkowski’s Tonal Patterns

 

See the bird, in the tree; See it ﬂy, 0- ver me.

 

Look up now, in the sky; there it goes, ﬂy—ing by.

After the two class periods, during which students had the opportunity to practice
the patterns, the students were brought individually into a familiar, private room. Each
student’s performance was audio-recorded as they vocally echoed the music teacher
playing the patterns they learned in class on the piano and then singing them. A Shure
SM58 was used to record the student’s performance. All performances were recorded
into Digidesing Mbox with the ProTools Software and burned onto a cd. Later, their
performances were rated using Rutkowski’s SVDM rating scale by two independent
judges who are experienced with children’s voices, as well as by the researcher to allow
for more meaningful interpretation of the results. Rutkowski’s scale follows.

1 “Pre-singer" does not sing but chants the song text.

1.5 "Inconsistent Speaking Range Singer " sometimes chants, sometimes sustains
tones and exhibits some sensitivity to pitch but remains in the speaking voice
range (usually A2 to C3).

2 "Speaking Range Singer " sustains tones and exhibits some sensitivity to pitch
but remains in the speaking voice range (usually A2 to C3).

2.5 "Inconsistent Limited Range Singer " wavers between speaking and singing

38

voice and uses a limited range when in singing voice (usually up to F 3).

3 "Limited Range Singer" exhibits consistent use of limited singing range (usually
D3 to F 3).
3.5 "Inconsistent Initial Range Singer " sometimes only exhibits use of limited

singing range, but other times exhibits use of initial singing range (usually D3 to
A3).

4 "Initial Range Singer " exhibits consistent use of initial singing range (usually
D3 to A3).

4.5 "Inconsistent Singer" sometimes only exhibits use of initial singing range, but
other times exhibits use of extended singing range (sings beyond the register lift:
B3-ﬂat and above).

5 "Singer" exhibits use of consistent extended singing range (sings beyond the

register lift: B3-ﬂat and above).

During the same two class periods and for the class period that followed, the
music teacher, who is also the researcher, taught the song for the singing game “On a
Mountain Stands a Climber” by rote. Each class reviewed the song and played short
rounds of the singing game to allow children to become familiar and conﬁdent with both
the song used in the game as well as the procedures of the game. The microphone that
would be used in recording the children's responses was used as part of the game during
this class period in an effort to help the children become accustomed to it. There was no
accompaniment used during the singing game. The music teacher demonstrated the
starting pitch of “On a Mountain Stands a Climber” with a ‘ready-sing’ before the class
began singing. The music teacher allowed the soloist to choose the next soloist after they
had sung. Following in Figure 2 is the singing game used in this study. The class sings

the beginning of the song together. While one child sings the solo by themselves,

39

inserting the name of the next child to sing the solo. Then, that same child “goes out to

play” for the next round of the song with assorted toys in a toy box.

Figure 2 — On a Mountain Stands a Climber _ Traditional

 

On a moun-tain stands a clim-ber. Who it is I do not know.

 

All he wants is fun ad-ven-ture. All he wants is a ﬁne voune friend. So

 

Come in dear while I go out to play.

Beginning in the third class period, the children were audio-recorded while
playing the game during their regular music class. A Crown SASS-P stereo recording
microphone was used to record the entire class while singing the song for the singing
game. A Shure SM58 was used to record the individual voice of the student singing the
solo. All performances were recorded into Digidesing Mbox with the ProTools Software
and burned onto a cd. Later, the same judges rated the singing game recorded responses
using the rating scale developed by Rutkowski as part of SVDM.

The judges received a packet of materials containing a copy each of Rutkowski’s
echo patterns, the Singing Game, the Singing Voice Development Measure rating scale, a
recording of the performances of all seventy-eight children for both the echo patterns and
the singing game, a score sheet for both the echo patterns and the singing game, and
practice examples of performances for use in training the judges to use SVDM. The
practice examples were heard ﬁrst by the judges to help them become familiar with

scoring using SVDM. Both judges were familiar with Rutkowski’s measure but had

40

never used it in practice. After the judges were comfortable using the scale, they scored
the recorded performances in one continuous sitting to avoid changes in scoring from one

day to the next.

41

CHAPTER FOUR
RESULTS AND INTERPRETATIONS

Reliabilities

After all ratings had been collected, Pearson Product-moment correlations were
performed to determine the inter-j udge reliability between the two independent judges.
Inter-judge reliability for Rutkowski’s SVDM was .84, showing a strong agreement
between the judges. Inter-judge reliability for the scores from the singing game was .90,
also showing a strong agreement between the judges. The inter-judge reliability was
higher for the singing game than for SVDM. The material performed by the child in the
singing game was not as long and varied as the echo patterns used in SVDM. One might
think that having more opportunity to assess a child’s voice would result in greater
agreement, but this was not the case in this study. A potential reason that the inter-judge
reliability was higher for the singing game than for the echo patterns could be that the
judges had more experience using the rating scale when rating singing game
performances, which were rated second. Another reason could be judge fatigue. Even
though the judges scored the singing game performances after the echo patterns, the time
it took to score the singing game was considerably shorter than the time used in scoring

the echo patterns.

Means and Standard Deviations

As seen in Table 1 below, Judge One had a mean score of 4. 1 86 with a standard
deviation of .944 for Rutkowski’s echo patterns and a mean score of 4.071 with a
standard deviation of 1.053 for the Singing Game. Judge Two had a mean score of 3.724
with a standard deviation of 1.175 for Rutkowski’s echo patterns and a mean score of
3.769 with a standard deviation of 1.136 for the Singing Game. Both judges showed

consistency from test to test in the scores they assigned the children’s performances.

42

This suggests the validity of using this singing game when assessing young children’s

singing voice development.

Table 1 — Means and Standard Deviations

 

 

 

 

 

 

 

 

Judge Mean Standard Deviation
Judge One Echo Patterns 4.186 .944
Judge Two Echo Patterns 3.724 1.175
Judge One Singing Game 4.071 1.053
Judge Two Singing Game 3.769 1.136

 

Correlation as a Measure of Concurrent Validity

Next, a Pearson Product-moment correlation was performed between the scores of

Rutkowski’s SVDM and the scores of the singing game to see if there was a meaningful

relationship between the two tests administered to assess singing voice development, thus

establishing concurrent validity. The correlation was .74. This is moderately high. This

singing game can be seen as a valid way to assess children’s singing voice development,

as demonstrated by the strong relationship between scores on SVDM and those of the

singing game.

43

 

CHAPTER FIVE
CONCLUSIONS AND RECOMMENDATIONS
Purpose
With the intent of providing a valid assessment tool for practicing elementary music
teachers, the purpose of this research was to investigate the measurement of singing voice

development through singing games.

Problem
The problem of this study was to determine the validity of using children’s singing games
as a context through which to measure singing voice development in young children,

using the rating scale of Rutkowski’s Singing Voice Development Measure.

Summary

Rutkowski’s Singing Voice Development Measure (SVDM) has proven to be a
valid form of assessing young children’s singing voice development (Rutkowski, 1990,
1996; Levinowitz et a1, 1998; Rutkowski & Miller, 2003). While her measure is valid
and useful to music education as a whole, it is not practical for the practicing general
music teacher. It requires taking children out of class to test them individually. This is
not practical when a music teacher has hundreds of children for which she is musically
responsible.

Through play, children explore and learn best. In the natural environment of play,
children are released from inhibitions. It was the hypothesis of the author that, by using a
singing game as a context in which to assess children’s singing voice development, one
might get an accurate measure of a child’s singing voice development.

The subjects were 78 third grade children from a suburban elementary school.
The children met two times a week for half an hour each for two consecutive weeks with

the music specialist at Lincoln Elementary School. The music specialist taught them

44

Rutkowski’s echo patterns from SVDM as well as the singing game “On a Mountain
Stands 3 Climber.” In the beginning of the third week, the children’s performances were
recorded. First, children were administered SVDM, a standardized test of singing voice
deve10pment. The children were taken out of class separately and tested in the music
room with the music specialist. Then, they played a singing game in their music class
that required the children to sing alone in the context of their regular music class. Their
solo performances were recorded and scored by two independent judges.

Then, the scores from SVDM were correlated with the scores from the singing
game to determine whether singing voice development could be measured with validity
in a singing game context. Inter-j udge reliability was high for both ratings from
Rutkowski’s echo patterns and ratings from the singing game. When scores from the two
tests were correlated, a strong positive relationship was found. This provided evidence of
the concurrent validity of using singing games as a context for measuring young
children’s singing voice development. This provides the practicing general music teacher
a means of measuring children’s singing voice development that is both practical and

developmentally appropriate.

Future Research

Though this study provided evidence that singing games can serve as a valid
context for measuring students’ singing voice development, further research should be
done on this subject to help support these results. Can other singing games be used for
the same purpose? This study should be replicated using other singing games.

In addition, more research should be done on singing games in general. Most
singing games today have a narrow melodic range due to the child’s deve10ping voice.
While these singing games should be used and embraced by the practicing music teacher,
there is a need for singing games that have a range that extends over the register lift so

that the general music teacher can measure the full range of their students’ singing voice

45

development. It is widely accepted that children learn best in their natural environment
of play. Yet, only two studies exist to date that focus on singing games; Brophy’s 1997
study and this one.

Rutkowski has contributed important research toward the understanding of
singing voice development, but many need to join her in this endeavor. Although many
researchers have studied intonation and other factors in singing, few have studied young
child’s singing voice development. Children must ﬁrst ﬁnd their singing voices before
they can begin to work on other singing qualities, such as intonation. Further research in
this area in crucial.

Teachers need useful and valuable tools to use in the general music classroom
when measuring student learning. Teachers and researchers need to join forces to
develop more practical tools to use in the classroom so that children can be taught to

achieve their fullest musical potential.

46

REFERENCES

About Kodaly. (n.d.). Retrieved August 24, 2004, from
http://www.oake.org/chapters1/keep/kodaly.html

Addison, R. (1991). Music and play. British Journal of Music Education. 8, 207-217.
Beaven, J. (2000). Joyful play. Montessori Life, 12, 38-41.

Berger, A., & Cooper, S. (2003). Musical play: A case study of preschool children and
parents. Journal of Research in Music Education, 51, 151-165.

Bredekamp, S., & Copple, C. (Eds). (1997). Developmentally appropriate practice in
early childhood programs serving children from birth through age 8 (Rev. ed.).
Washington DC: National Association for the Education of Young Children.

Brophy, T. S. (1992). Play and the young child: Musical implications. Gainesville, FL:
The University of Florida. (ERIC Document Reproduction Services No. ED
358934)

Brophy, T. S. (1997). Authentic assessment of vocal pitch accuracy in ﬁrst through third
grade children. Contributions to Music Education, 24, 57-70.

Byo, S. J. (2000). Classroom teachers’ and music specialists’ perceived ability to

implement the national standards for music education. Arts Education Policy
Review, 101, 30-35.

Cronbach, L. J. (1970). Essentials of psychological testing (3rd Ed). New York,
Evanston, London: Harper and Row, Publishers.

Csikszentmihalyi, M. (2000). Beyond boredom and anxiety. San Francisco, CA: Jossey-
Bass Publishers.

Dalby, M. (1992). Teaching musicianship with games. The Instrumentalist, 21-24,56.

Elliott, D. (1995). Music matters a new philosophy of music education. New York:
Oxford University Press.

Feirabend, J. M. (1996). Music and movement for infants and toddlers: naturally
wonder-full [Electronic version]. Early Childhood Connebtions, 2, 1-10.

Feierabend, J. M. (1998). The Hartt school’s national center for music in the early years
[Electronic version]. Early Childhood Connections, 4, 1-4.

Feierabend, J. M. (2000). First steps in music for early elementary. Chicago: GIA

47

Publications, Inc.

Friedrich F roebel: Founder, ﬁrst kindergarten. (2000). Scholastic Early Childhood Today.
15, 63.

Frost, J. L. (1998). Neuroscience, play, and child development. Austin, TX: University of
Texas at Austin. (ERIC Document Reproduction Services No. ED 427845)

Gagne, D. (1997). Singing games children love (1" ed). Alberta: Themes and Variations.

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, 93-99.

Goetze, M., Cooper, N., & Brown, C. J. (1989). Recent research on singing in the general
music classroom. Bulletin of the Council for Research in Music Education, 3 7,
16-37.

Goetze, M., & Horii, Y. (1989). A comparison of the pitch accuracy of group and
individual singing in young children. Bulletin of the Council for Research in

Music Education, 99, 57-73.

Gordon, E. (1971). The psychology of music teaching. Englewood Cliffs, NJ: Prentice
Hall, Inc.

Gordon, B. (2001). Preparatory audiation, audiation, and music learning theory.
Chicago, IL: GIA Publications.

Gordon, E. (2003). Learning sequences in music. Chicago, IL: GIA Publications.

Gould, A. O. (1969). Developing specialized programs for singing in the elementary
school. Bulletin of theCouncil for Research in Music Education, 17, 9-22.

Green, G. A. (1989). The effect of vocal modeling on pitch-matching accuracy of
elementary schoolchildren. Journal of Research in Music Education, 38, 225-231.

Hotchkiss, G., & Athey, M. (1978). Music learning grows with games. Music Educators
Journal, 64, 48-51.

Howle, M. J. (1997). Play-party games in the modern classroom. Music Educators
Journal, 83, 24-28.

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, 23 7-246.

48

Kratus, J. (1997). The roles of work and play in music education. Paper presented at the
Philosophy of Music Education International Symposium III, Los Angeles, CA.

Levinowitz, L., Barnes, P., Guerrini, S., Clement, M., D’April, P., & Morey, M. (1998).
Measuring singing voice development in the elementary general music classroom.
Journal of Research in Music Education, 46, 35-47.

McGuire, K. (2002). Doing it all: linking the elements of music and the standards. Music
Educators Journal, 89, 49-52.

Mitchell, L. (1991). One, two, three...Echo me! New York: Parker Publishing Company

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, 42,5-13.

Music Educators National Conference (1994). What every young American should know
and be able to do in the arts: National standards for arts education. Reston, VA:
Author.

Piaget, J. (1962). The stages of the intellectual development of the child. Bulletin of the
Menninger Clinic, 26, 120-128.

Prim, F. M. (1995). Tradition and change in children’s games: Its implication in music
education. Bulletin of the Council for Research in Music Education, 127,
149-154.

Rutkowski, J. (1990). The measurement and evaluation of the children’s singing voice
development. The Quarterly Journal of Music Teaching and Learning, 1, 81-95.

Rutkowski, J. (1996). The effectiveness of individual/small-group singing activities on
kindergartners’ use of singing voice and developmental music aptitude. Journal of
Research in Music Education, 44, 353-368.

Rutkowski, J ., & Miller, M. (2003). A longitudinal study of elementary children’s
acquisition of their singing voices. Update: Applications of Research in Music
Education, 22, 5-14.

Rutkowski, J. (2004, March). Helping all children learn to sing. Paper presented at
meeting at Michigan State University, Lansing, MI.

Shamrock, M. (1997). Orff-Schulwerk an integrated foundation. Music Educators
Journal, 83, 41-44.

Taggart, C. (2003). Child-centered play in music: Developmentally appropriate practice.
Early Childhood Connections, 9, 15-23.

49

Van Der Linde, C. (1999). The relationship between play and music in early childhood:
Educational insights. Education, 119, 610-615.

Welch, G. (1994). The assessment of singing. Psychology of Music and Music Education,
22, 3-19.

Yarbrough, C., Bowers, J ., & Benson, W. (1992). The effect of vibrato on the pitch-
matching accuracy of certain and uncertain singers. Journal of Research in Music
Education, 40, 30-38.

Yarbrough, C., Bowers, J ., Benson, W., & Green, G. (1991). Inaccurate singers: An
exploratory study of variables affecting pitch-matching. Bulletin of the Council
for Research in Music Education, 107, 23-34.

Yurko, M. (1992). Music mind games (Revised ed). Miami, FL: Warner Bros.
Publications.

50

MICHIGAN STATE UNIVERSITY LIB

l3 3||| || |3| |||3||||||||Es

0662

||||3 ||1