DEVELOPMENT AND VALIDATION OF A MUSICAL BEHAVIOR MEASURE
FOR PRESCHOOL CHILDREN
By
Gina Jisun Yi

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Music Education - Doctor of Philosophy
2013

ABSTRACT
DEVELOPMENT AND VALIDATION OF A MUSICAL BEHAVIOR MEASURE
FOR PRESCHOOL CHILDREN
By
Gina Jisun Yi
The purpose of this study was to develop a measure for use in assessing musical
behaviors of preschool children in the context of regular music instruction and to determine the
validity and the reliability of the measure. The Early Childhood Musical Behavior Measure
(ECMBM) was constructed for use with preschool-aged children to measure their musical
behaviors during music class. ECMBM is a rating scale with 12 continuous dimensions for use
in assessing the following: Tonal Skills (Singing Accuracy, Resting Tone, Major Pattern
Imitation, Minor Pattern Imitation); Rhythm Skills (Chant Performance, Duple Pattern Imitation,
Triple Pattern Imitation, Rhythm Improvisation, Keeping Beat: Gross Motor, Keeping Beat:
Manipulatives, Keeping Beat: Locomotor); and Movement.
I conducted a pilot study to practice and refine the assessment techniques and instruments
that I would be using in the main study and to determine the preliminary reliabilities of the scale.
Also, the measure was checked for content validity both subjectively and objectively with seven
university professors who are experts in early childhood music education and music education
measurement. Based on the results of the pilot study, comments from the judges in the pilot
study, and comments from validation panel members, ECMBM was revised for use in the main
study.
To determine the reliability of the measure, the musical behaviors of 4- and 5-year-old
children in two intact classes were audio and/or video recorded, and three independent judges
rated the children’s musical behaviors. These ratings were correlated to determine interjudge

reliabilities. Also, I correlated my in-class ratings of the children’s performances with my after
class ratings from listening and watching to the recorded data to obtain the intrajudge reliabilities
and to determine whether the ratings that I made during class were sufficiently reliable.
On the basis of the results of this study, it is reasonable to state that early childhood
music teachers who use a Music Learning Theory curriculum can use ECMBM to assist them in
assessing preschool children’s musical behaviors in the context of music instruction. The
intrajudge and interjudge reliabilities indicate that the Tonal Behavior rating scales are reliable
and Rhythm Behavior rating scales also are reliable except for the Chant Performance dimension.
The content validation process as well as the intercorrelations provided evidence that they also
are valid. The Movement dimension lacked content validity, and its ratings also resulted in low
to moderate reliabilities.
There is a need for further research that pertains to the refinement of the rating scales and
assessment activities for music instruction at all levels. In addition, this study should be
replicated with a larger sample, different group of children in terms of socio-economic status and
racial make up, and with different criterion songs, chants, or patterns to investigate further the
validity of ECMBM.

	
  

Copyright by
GINA JISUN YI
2013

	
  

iv	
  

This dissertation is dedicated to my mother and father.
To mom: Thank you for setting an example on how to live a life and to endure through
perseverance, for being my first great music teacher, and for just being my mom.
To dad: Thank you for providing the best education possible, believing in me, supporting me,
and for being such a good father. And most of all, encouraging me to “go for the dream.”

	
  

v	
  

ACKNOWLEDGEMENTS

I owe a great debt of thanks to many people and show my sincere appreciation, as this
journey could not have been possible without them. First, many sincere thanks to my committee
members: Dr. Mitchell Robinson, for your genuine support, care, and humor; Dr. Sandra Snow
for your valuable suggestions; Dr. Derek Polischuk for your friendship and encouragement. And
to my advisor and Chair, Dr. Cynthia Taggart for endless inspiring guidance, supporting and
encouraging me and challenging me to be a better scholar.
Thank you to Carla Larzelere, Carin McEvoy, and Heather Shouldice, for your insights
on early childhood music and participation as raters. And thank you to Julie Kastner and Karen
Salvador for participation as raters, your friendship, and thoughts on music education.
Special thanks to Dr: Suzanne Burton, Denise Guilbault, Lisa Koops, Herbert Marshall,
Jill Reese, Joanne Rutkowski, and Wendy Valerio for their participation as members for
validation process. Deepest thanks to all of my Pre-K students who participated in this study.
Thank you for having fun during data collection process and your precious smiles. Special
thanks to the Pre-K teachers: Emma Lewis and Regina Bailey for their willingness to
accommodate me and supporting me during my study.
Finally, I give my thanks and praise to my Lord and Savior, Jesus Christ.
The Lord will guide you always; he will satisfy your needs in a sun-scorched land and
will strengthen your frame. You will be like a well-watered garden, like a spring whose
waters never fail. (Isaiah 58:11)
Throughout the journey, I was blessed for His guidance, His wisdom, and His flowing words of
comfort. To Him be the glory!

	
  

vi	
  

TABLE OF CONTENTS

LIST OF TABLES...........................................................................................................................x
LIST OF FIGURES ..................................................................................................................... xiii
CHAPTER I
INTRODUCTION ...........................................................................................................................1
Current Trends in Assessment ....................................................................................................2
Accountability........................................................................................................................2
Teacher Evaluation ................................................................................................................4
Issues in Assessing Young Children...........................................................................................5
Developmentally Appropriate Assessment ...........................................................................6
Authentic Assessment ...........................................................................................................7
Assessment in the Musical Domain ......................................................................................8
Qualitative and Quantitative Assessment ...........................................................................10
Purpose and Research Questions ..............................................................................................11
Delimitations.............................................................................................................................11
CHAPTER II
REVIEW OF LITERATURE ........................................................................................................13
Music Learning in Preschool Settings ......................................................................................13
Musical Behaviors of Preschool Children ................................................................................14
Singing Behavior .................................................................................................................17
Movement Behavior ............................................................................................................18
Rhythmic Behavior ..............................................................................................................21
Instrument Playing ...............................................................................................................23
Creativity and Improvisation ...............................................................................................24
Assessing Musical Behavior of Preschool Children ................................................................25
Theoretical Frames ..............................................................................................................25
Measurement Tools..............................................................................................................26
Checklists or Additive Rating Scales..............................................................................27
Numerical Rating Scales or Rubrics ...............................................................................28
Continuous Rating Scales ...............................................................................................30
Singing ...................................................................................................................31
Pattern Performance...............................................................................................32
Movement and Rhythm .........................................................................................33
CHAPTER III
METHODOLOGY ........................................................................................................................35
Subjects .....................................................................................................................................35
Development of the Musical Behavior Measure ......................................................................36
Characteristics of a Musical Behavior Measure ..................................................................36
Development Procedures .....................................................................................................37

	
  

vii	
  

Construction of Dimensions and Criteria ............................................................................38
Construction of Assessment Activities ................................................................................41
Description of Assessment Activities ..................................................................................42
Tonal Behavior: Singing Accuracy.................................................................................42
Tonal Behavior: Resting Tone ........................................................................................43
Tonal/Rhythm Behavior: Tonal/Rhythm Pattern Imitation ............................................44
Rhythm Behavior: Chant Performance...........................................................................46
Rhythm Behavior: Rhythm Improvisation .....................................................................47
Rhythm Behavior: Keeping Beat ....................................................................................47
Movement Behavior .......................................................................................................49
Design and Procedures..............................................................................................................50
Data Collection ....................................................................................................................50
Rating Process......................................................................................................................51
Data Analysis .......................................................................................................................51
CHAPTER IV
THE PILOT STUDY .....................................................................................................................53
Findings ....................................................................................................................................54
Audio and Videotaping Procedures .....................................................................................54
Piloting Assessment Activities ............................................................................................56
Interjudge Reliabilities.........................................................................................................58
Tonal Behaviors ..............................................................................................................58
Rhythm Behaviors ..........................................................................................................61
Movement Behavior .......................................................................................................66
Means and Standard Deviations ..........................................................................................66
Discussion of the Pilot Study....................................................................................................71
CHAPTER V
FURTHER REVISIONS OF THE RATING SCALE...................................................................75
Content Validation of the ECMBM .........................................................................................75
Sequencing of the Criteria ...................................................................................................76
Panel Members’ and Judges’ Comments.............................................................................78
Tonal Behavior: Singing Accuracy.................................................................................78
Tonal Behavior: Resting Tone ........................................................................................80
Tonal Behavior: Tonal Pattern Imitation ........................................................................81
Rhythm Behavior: Chant Performance...........................................................................83
Rhythm Behavior: Rhythm Pattern Imitation .................................................................84
Rhythm Behavior: Rhythm Improvisation .....................................................................86
Rhythm Behavior: Keeping Beat ....................................................................................88
Movement Behavior .......................................................................................................89
CHAPTER VI
RESULTS ......................................................................................................................................93
Reliabilities ...............................................................................................................................93
Intrajudge Reliabilities.........................................................................................................93
Interjudge Reliabilities.........................................................................................................96

	
  

viii	
  

Means, Standard Deviations, and Standard Errors of Measurement of Final ECMBM...........98
Intercorrelations of Dimensions..............................................................................................100
Interpretations and Discussions ..............................................................................................102
Comparison of the Results of the Pilot and Main Study....................................................102
Intrajudge Reliabilities.......................................................................................................104
Discussion ...............................................................................................................................105	
  
Rhythm Behavior: Chant Performance..............................................................................105
Movement Behavior ..........................................................................................................106
Intercorrelations .................................................................................................................107
CHAPTER VII
SUMMARY, IMPLICATIONS, RECOMMENDATIONS, AND CONCLUSION ..................109
Summary .................................................................................................................................109
Purpose and Research Questions .......................................................................................109
Design and Analysis ..........................................................................................................109
Results and Discussions.....................................................................................................111
Implications ............................................................................................................................114
Recommendations for Future Research ..................................................................................115
Conclusion ..............................................................................................................................116
APPENDICES .............................................................................................................................118
Appendix A: Teacher Consent Form ......................................................................................119
Appendix B: Parent Consent Form .........................................................................................121
Appendix C: Child Assent Form ............................................................................................123
Appendix D: Validation Panel Members................................................................................124
Appendix E: Validation Forms ...............................................................................................125
Appendix F: Instructions and Rating Sheets...........................................................................132
Appendix G: Rating Cards......................................................................................................142
Appendix H: Criterion Songs and Chant ................................................................................148
REFERENCES ............................................................................................................................151

	
  

ix	
  

LIST OF TABLES

Table 1.1

Pearson Correlations among Judges’ Composite Scores of All Dimensions ........58

Table 1.2

Pearson Correlations among Judges’ Composite Scores of Tonal Behavior
Dimensions ............................................................................................................59

Table 1.3

Pearson Correlations among Judges’ Ratings on Singing Accuracy Dimension ..59

Table 1.4

Pearson Correlations among Judges’ Ratings on Resting Tone Dimension..........60

Table 1.5

Pearson Correlations among Judges’ Ratings on Major Pattern Imitation
Dimension ..............................................................................................................60

Table 1.6

Pearson Correlations among Judges’ Ratings on Minor Pattern Imitation
Dimension ..............................................................................................................60

Table 1.7

Pearson Correlations among Judges’ Composite Scores of Tonal Pattern Imitation
Dimensions ............................................................................................................61

Table 1.8

Pearson Correlations among Judges’ Composite Scores of Rhythm Behavior
Dimensions ............................................................................................................61

Table 1.9

Pearson Correlations among Judges’ Ratings on Chant Performance Dimension
...............................................................................................................................62

Table 1.10

Pearson Correlations among Judges’ Ratings on Duple Pattern Imitation
Dimension ..............................................................................................................63

Table 1.11

Pearson Correlations among Judges’ Ratings on Triple Pattern Imitation
Dimension ..............................................................................................................63

Table 1.12

Pearson Correlations among Judges’ Composite Scores of Rhythm Pattern
Imitation Dimensions.............................................................................................63

Table 1.13

Pearson Correlations among Judges’ Ratings on Rhythm Improvisation
Dimension ..............................................................................................................64

Table 1.14

Pearson Correlations among Judges’ Composite Scores of Keeping Beat
Dimensions ............................................................................................................64

Table 1.15

Pearson Correlations among Judges’ Ratings on Keeping Beat: Gross Motor
Dimension ..............................................................................................................65

	
  

x	
  

Table 1.16

Pearson Correlations among Judges’ Ratings on Keeping Beat: Manipulatives
Dimension ..............................................................................................................65

Table 1.17

Pearson Correlations among Judges’ Ratings on Keeping Beat: Locomotor
Dimension ..............................................................................................................66

Table 1.18

Pearson Correlations among Judges’ Ratings on Movement Behavior Dimension
...............................................................................................................................66

Table 2.1

Means and Standard Deviations of Composite Scores of All Dimensions by
Judges.....................................................................................................................67

Table 2.2

Means and Standard Deviations of Tonal Behavior Dimensions ..........................68

Table 2.3

Means and Standard Deviations of Rhythm Behavior Dimensions ......................68

Table 2.4

Means and Standard Deviations of Duple and Triple Pattern Imitation Dimensions
...............................................................................................................................69

Table 2.5

Means and Standard Deviations of Keeping Beat Dimensions by Mode..............70

Table 2.6

Means and Standard Deviations of Rhythm Improvisation Dimension ................70

Table 2.7

Means and Standard Deviations of Movement Behavior Dimension....................71

Table 3.1

Correct Responses for Singing Accuracy Dimension............................................76

Table 3.2

Correct Responses for Resting Tone Dimension ...................................................76

Table 3.3

Correct Responses for Chant Performance Dimension .........................................77

Table 3.4

Correct Responses for Rhythm Improvisation Dimension ....................................77

Table 4.1

Intrajudge Reliabilities, Means, Standard Deviations, and Standard Errors of
Measurement for Individual Dimensions ..............................................................94

Table 4.2

Theoretical and Observed Means, Standard Deviations, and Standard Errors of
Measurement for Combined Dimensions ..............................................................95

Table 5.1

Pearson Correlations for Individual Tonal Behavior Dimensions.........................96

Table 5.2

Pearson Correlations for Individual Rhythm Behavior Dimensions .....................96

Table 5.3

Pearson Correlations for Composite Scores of Combined Dimensions ................97

	
  

xi	
  

Table 6.1

Means, Standard Deviations, and Standard Errors of Measurement of Individual
Dimensions by Judges ...........................................................................................98

Table 6.2

Theoretical and Observed Means and Standard Deviations of Composite Scores
by Judges .............................................................................................................100

Table 7

Intercorrelations of the 12 Dimensions Using the Judges’ Combined Ratings of
Individual Dimensions .........................................................................................101

	
  

xii	
  

LIST OF FIGURES

Figure 1

The Early Childhood Musical Behavior Measure .................................................39

Figure 2

Criterion Song for Resting Tone Dimension .........................................................44

Figure 3

Tonal and Rhythm Patterns Used for Tonal and Rhythm Pattern Imitation
Dimensions ............................................................................................................45

Figure 4

Criterion Chant for Chant Performance Dimension ..............................................46

Figure 5

Preparing Chant for Chant Performance Dimension .............................................46

Figure 6

Teacher-Initiated Patterns for Rhythm Improvisation Dimension ........................47

Figure 7

Criteria 3 and 4 of the Original Rhythm Improvisation Dimension ......................77

Figure 8

The Original and Revised Rating Scales for Singing Accuracy Dimension .........79

Figure 9

The Original and Revised Rating Scales for Resting Tone Dimension.................81

Figure 10

The Original and Revised Rating Scales for Tonal Pattern Imitation Dimension
...............................................................................................................................82

Figure 11

The Original and Revised Rating Scales for Chant Performance Dimension .......84

Figure 12

The Original and Revised Rating Scales for Rhythm Pattern Imitation Dimension
...............................................................................................................................85

Figure 13

The Original and Revised Rating Scales for Rhythm Improvisation Dimension
...............................................................................................................................87

Figure 14

The Original and Revised Rating Scales for Keeping Beat Dimension ................88

Figure 15

The Original and Revised Rating Scales for Movement Behavior Dimension .....91

Figure 16

Smiley Faces in Child Assent Form ....................................................................123

Figure 17

Criterion Tonal Patterns for Tonal Pattern Imitation Dimensions.......................127

Figure 18

Criterion Rhythm Patterns for Rhythm Pattern Imitation Dimensions................128

Figure 19

Teacher-Initiated Rhythm Patterns for Rhythm Improvisation Dimension.........129

	
  

xiii	
  

Figure 20

Movement Element Chart ....................................................................................131

Figure 21

A Rating Sheet for Judges to Record Children’s Singing Accuracy Scores .......133

Figure 22

Criterion Song for Singing Accuracy Dimension................................................133

Figure 23

A Rating Sheet for Judges to Record Children’s Resting Tone Scores...............134

Figure 24

Criterion Tonal Patterns for Tonal Pattern Imitation Dimensions.......................135

Figure 25

A Rating Sheet for Judges to Record Children’s Tonal Pattern Imitation Scores
.............................................................................................................................135

Figure 26

A Rating Sheet for Judges to Record Children’s Chant Performance Scores .....136

Figure 27

Criterion Chant for Chant Performance Dimension ............................................137

Figure 28

A Rating Sheet for Judges to Record Children’s Rhythm Pattern Imitation Scores
.............................................................................................................................137

Figure 29

Criterion Rhythm Patterns for Rhythm Pattern Imitation Dimensions................138

Figure 30

Teacher-Initiated Rhythm Patterns for Rhythm Improvisation Dimension
.............................................................................................................................138

Figure 31

A Rating Sheet for Judges to Record Children’s Rhythm Improvisation Scores
.............................................................................................................................139

Figure 32

A Rating Sheet for Judges to Record Children’s Keeping Beat Scores ..............140

Figure 33

A Rating Sheet for Judges to Record Children’s Movement Scores ...................141

Figure 34

A Rating Card Used for Recording Children’s Singing Accuracy Scores ..........142

Figure 35

A Rating Card Used for Recording Children’s Resting Tone Scores..................142

Figure 36

A Rating Card Used for Recording Children’s Tonal Pattern Imitation Scores..143

Figure 37

A Rating Card Used for Recording Children’s Chant Performance Scores ........144

Figure 38

A Rating Card Used for Recording Children’s Rhythm Pattern Imitation Scores
.............................................................................................................................145

Figure 39

A Rating Card Used for Recording Children’s Rhythm Improvisation Scores...146

Figure 40

A Rating Card Used for Recording Children’s Keeping Beat Scores .................146

	
  

xiv	
  

Figure 41

A Rating Card Used for Recording Children’s Movement Scores......................147

Figure 42

Criterion Songs and Chant Used for Assessment Activities................................148

	
  

xv	
  

CHAPTER I
INTRODUCTION
Bjørkvold (1992) coined the term muse within to refer to humans’ natural tendency to be
engaged in music as muse-ical beings. He remarked, “We all need the Muse Within, for we are
what I shall call muse-ical beings. To lose our muse-icality would be to lose a profoundly
essential part of our humanity” (Bjørkvold, 1992, p. xiii). Humans, as muse-ical beings, exhibit
musicality through various musical behaviors from infancy. From infants’ intuitive behaviors attentive listening, turning their heads toward a musical source, waving their arms, kicking their
legs or sucking in response to musical sound - to preschool children’s intentional behaviors and
beyond, children make and respond to music.
Beginning in the late 1960s, there was an upsurge of interest in early childhood music
education (Greenberg, 1976; Simons, 1986). Children’s musical behaviors (Miller, 1986; Moog,
1976; Moorhead & Pond, 1978), the development of musical abilities (Jersild & Bienstock, 1934,
1935; Rutkowski, 1990; Shelley, 1981; Simons, 1986), and music achievement (Blesedell, 1991;
Taggart, 1994) in early childhood have been investigated both qualitatively and quantitatively.
In addition, pedagogues turned their attention to music in early childhood (Alvarez, 1981;
Gordon, 2013; Jaques-Dalcroze, 1967; Levinowitz & Guilmartin, 2000; Nash, 1974) and to early
childhood music curricula. Consequently, writings on the music learning of young children have
proliferated. However, assessing children’s music learning has been generally disregarded, as
the notion of assessing the child’s personal “aesthetic-expression” does not seem to correspond
with developmentally appropriate, child-centered practice and process-oriented learning
(Rauscher, 2000; Wright, 1994).

	
  

1	
  

Yet, Copple and Bredekamp (2009) contend, “Teachers cannot be intentional about
helping children to progress unless they know where each child is with respect to learning goals”
(p. 22). As such, assessment and learning are linked, because assessment enables teachers to not
only plan and evaluate curriculum but also to make instruction simpler or more complex
according to the readiness and needs of individual children (Copple & Bredekamp, 2009).
Similarly, researchers and measurement experts in music education emphasize the importance of
assessing students’ musical achievement, as assessing enables teachers to design appropriate
lesson plans for and to evaluate their instruction (Abeles, Hoffer, & Klotman, 1994; Boyle, 1989).
Current Trends in Assessment
Accountability
While many early childhood teachers informally monitor and assess children’s behavior
during class, relatively few teachers systematically observe, evaluate, and document children’s
learning and growth (National Research Council [NRC], 2001). However, with increasing
demand for school and teacher accountability from the public, parents, and government,
systematic documentation that offers some objectivity in assessing children is becoming
necessary.
Student learning can be documented through various means of formal assessment and/or
standardized testing. Accountability became an important issue as standards for K-12 education
were developed, and states developed their own versions of those standards. Assessment
guidelines to determine if children were meeting those standards accompanied the position
statements offered by various professional education organizations (National Association for the
Education of Young Children & National Association of Early Childhood Specialists in State
Departments of Education [NAEYC & NAECS/SDE], 2003; National Association for Music

	
  

2	
  

Education [NAfME], 2012a). Furthermore, the No Child Left Behind Act of 2001 (NCLB)
mandated assessment and accountability at all levels of public schools, even in preschool (Fiore,
2012; NRC, 2001). In reaction to NCLB legislation, many public schools began administering
standardized tests or assessments in order to provide evidence of student achievement. While the
intention of NCLB was to systematically impact public education and raise the academic
standards of all children so that not a single child would be overlooked (Ravitch, 2010), many
educators concur that the result was not successful, and the pressures put on teachers to raise the
standards of their students resulted in teaching to the test and in children spending time learning
to regurgitate the test’s content in class on a daily basis (Miller & Almon, 2009). Simply,
optimal learning was not occurring at school partly as a result of standardized testing, and, most
important, developmentally appropriate learning for young children was put in jeopardy (Fiore,
2012; Miller & Almon, 2009; Puckett & Black, 2000).
Some policy makers advocate standardized testing because they believe that it offers
objectivity. However, while “objective measures are designed to ensure that there is as little bias
and variability as possible to uphold scientific standards” (Fiore, 2012, p.16), no measure is
completely objective, because any test also contains some bias from the person constructing the
test - choosing the form of questions (i.e., multiple choice/true or false) being asked, range of
possible responses, and determining which one is correct (Fiore, 2012). Educators also express
concerns about the appropriateness of standardized testing for young children. Many early
childhood experts contend that testing is problematic for young children because it does not give
holistic information about complex, developing children, and children are not represented well
by this type of testing (Fiore, 2012). Mainly, because standardized testing or structured
assessment represents an artificial setting, children tend to respond out of context and have

	
  

3	
  

difficulty understanding the instructions, situational clues, and other testing stimuli (NRC, 2001).
Because of the nature of this type of test, there is a consensus among early childhood educators
that standardized testing is invalid and not developmentally appropriate for children under age 8
(Miller & Almon, 2009). Therefore, in order to suggest an alternative for standardized testing
and to provide for accountability and appropriate communication with parents and the public
about students’ learning, the teachers’ role in documenting children’s learning and growth with
appropriate measurement or assessment methods becomes critical.
Teacher Evaluation
As a result of recent legislation concerning teacher effectiveness, teachers are being
asked to document student learning in new ways. Students’ growth is now considered a
“significant factor” in determining teacher pay and continued employment, so documentation of
student learning is required by many states (National Council on Teacher Quality [NCTQ], 2012).
While assessment at the state level most often occurs through standardized testing, education
organizations emphasize the importance of local assessments (Forum on Educational
Accountability [FEA], 2007; Michigan Education Association [MEA], 2011; NAEYC &
NAECS/SDE, 2003), so teachers’ abilities to assess and document student learning is critical.
In their 2012 annual report, NCTQ (2012) reported that 30 states require at least some
objective measure of student achievement as part of teacher evaluations, and 11 states make it
the most significant factor. Moreover, eleven states: Colorado, Delaware, Florida, Louisiana,
Maryland, Michigan, Nevada, Ohio, Oklahoma, Rhode Island, and Tennessee, make student
achievement growth the preponderant factor in evaluating teachers. And, of the 11 states, nine Colorado, Delaware, Florida, Louisiana, Michigan, Nevada, Oklahoma, Rhode Island, and
Tennessee - require objective evidence of students learning for tenure decisions. Consequently,

	
  

4	
  

teachers not only are being asked to document the learning of their students in a systematic way,
but this documentation will be central to their evaluations, will be a determining factor in their
tenure, and will determine which teachers will be retained when cuts occurs in the school.
Issues in Assessing Young Children
Standardized testing is present now even in kindergarten and could be passed on to
preschool (Copple & Bredekamp, 2009; Miller & Almon, 2009; NRC, 2001), and related survey
studies of school systems in big cities such as New York and Los Angeles reported that
standardized testing and preparation for tests are occurring in kindergarten as part of a daily
activity, with children spending hours each day in literacy, math, and test-preparation (Miller &
Almon, 2009). Assessment serves many purposes in early childhood education, such as
supporting learning, identification of special needs, program evaluation and monitoring trends,
establishing accountability (Shepard, Kagan, & Wurtz, 1998), and determining school readiness.
Yet arguably the most important purpose in assessing young children is to inform instruction so
that teachers can plan programs and curricula that promote each child’s learning. However,
standardized assessment, because it does not capture a holistic view of the child, is not valid for
this purpose and results in serious errors in terms of assessment (McAfee & Leong, 2002; Miller
& Almon, 2009). Therefore its use in improving instruction is limited.
Because early childhood is a dramatic period for development and learning in motor,
social-emotional, and cognitive development, it is challenging for teachers to capture children’s
abilities using any single means of assessment. Therefore, early childhood experts recommend
that assessment should not rely on a single measure or source of information; multiple sources
should be considered to increase the validity of an evaluation (Copple & Bredekamp, 2009;
McAfee & Leong, 2002; NAEYC & NAECS/SDE, 2003; NRC, 2001).

	
  

5	
  

During early childhood, children develop in the physical, cognitive, social, and emotional
domains (Beaty, 2010; Copple & Bredekamp, 2009; Guddemi, 2003) and in language and
creative development (Beaty, 2010). Regardless of the domain, assessment should align with
learning standards that are developmentally appropriate and must be embedded in classroom
activities (Copple & Bredekamp, 2009; Miller & Almon, 2009). That is, if a teacher wants to
assess literacy, that teacher must know how to align the reading and writing of the assessment so
that it is appropriate for the students. For example, children may pretend to read, turning pages
and making up stories with pictures, or matching some letters to a sound. For such students,
teacher observation or child interview would be more appropriate assessment methods than
standardized testing to evaluate what the child knows in relation to his or her current
developmental stage. Because of its stance that most forms of testing before age eight are
inappropriate, NAEYC has recommended since 1987 that children’s progress be documented
using developmentally appropriate practices and points toward use of non-standardized
assessments, including teacher observations and work assessment or portfolios (Miller and
Almon, 2009).
Developmentally Appropriate Assessment
The National Association for the Education of Young Children (NAEYC) and the
National Association of Early Childhood Specialists in State Departments of Education
(NAECS/SDE) proposed guidelines for practice in the assessment of young children (birth to age
8) in a joint position statement. They proposed that assessment methods should be
developmentally appropriate, culturally and linguistically responsive, tied to children’s daily
activities, supported by professional development, inclusive of families, and connected to
specific, beneficial purposes (NAEYC & NAECS/SDE, 2003). NAEYC further suggested that

	
  

6	
  

developmentally appropriate assessment for preschool children be: (1) strategic and purposeful,
(2) systematic and ongoing, (3) integrated with teaching and curriculum, (4) valid and reliable,
and (5) communicated and shared with parents or colleagues (Copple & Bredekamp, 2009).
Authentic Assessment
The misuse of standardized testing and the pressures to teach to the test have prompted
many early childhood teachers to consider using authentic assessment as a substitute for or in
addition to standardized testing (Fiore, 2012). Neisworth and Bagnato (2004) noted, “When
assessment is ‘authentic,’ it yields information about functional behavior in children’s
typical/natural settings, what they really know and do” (p. 203). Authentic assessment offers
vastly different characteristics from standardized testing in terms of (1) context; it is situated in
the natural environment, (2) content; it focuses on natural behaviors that have functional
importance to the child and their progress, (3) procedure; children are assessed through
observations, interviews, portfolios, and rating scales, and (4) the assessor; the assessor is a
familiar person to the child (Neisworth & Bagnato, 2004).
Authentic assessment meets many of the DAP criteria; authentic assessment is childcentered, process-oriented, performance-based, aims to solve realistic tasks, enables the teacher
to record, assess, and analyze a variety of behaviors of children, and, most important, occurs
within the context of children’s meaningful “doing” (Bagnato, 2007). Therefore, authentic
assessment is recommended highly for use with young children, for whom pencil and paper tests
are not valid measures (Fiore, 2012; NRC, 2001; Shepard et al., 1998).
Authentic assessment offers various windows into the child’s learning as their teachers
systematically collect information through observation (e.g., anecdotal notes), systematic
recording (e.g., checklist), interviews, work samples, and rating scales (Neisworth & Bagnato,

	
  

7	
  

2004). When teachers observe to assess children’s growth and learning, rating scales might be
used in measuring specific skills (Campbell & Scott-Kassner, 2006). Assessing and collecting
information in such a multidimensional way supplies rich information about children’s growth
and learning, and not only has immediate application in the classroom but also meets the
accountability standards set by the state and the NCLB legislation (Fiore, 2012).
Assessment in the Musical Domain
When assessing young children’s musical behaviors, in order to adhere to
developmentally appropriate practice, teachers should be certain that their students are familiar
with the context and the nature of the task, because children are reluctant to perform when they
are skeptical of their abilities, especially in an unfamiliar setting.
Because children learn by doing and best exhibit what they know through action,
observation is considered one of the best ways to assess the skills of young children and is
especially well-suited to preschool children (Copple & Bredekamp, 2009; Seefeldt, 1998).
Seefeldt (1998) remarked, “Young children reveal themselves through their behaviors. . . . [and]
observing them often yields accurate information” (p. 317). Because of that, early childhood
experts concur that the best assessment method is to observe young children in their regular
classrooms using a particular set of criteria that is based upon children’s natural and spontaneous
behaviors to guide observations (Beaty, 2010).
Similarly, early childhood music researchers have remarked that teachers often rely on
observation when assessing children’s musical development and behaviors (Etopio & Cissoko,
2005). Furthermore, the use of observation for investigating children’s musical behavior has
been prevalent in both qualitative (Barrett, 2006; Berger & Cooper, 2003; Burton, 2002;
Hornbach, 2005; Kierstead, 2006; Metz, 1989; Miller, 1986; Moorhead & Pond, 1978; Reynolds,

	
  

8	
  

1995; Shelley, 1981; Young, 2002, 2003), and quantitative research (Moog, 1976). Observation,
often accompanied by the use of measurement tools such as rating scales or checklists when
observing musical behavior in the context of music activities, provides rich information of
children’s musical behavior.
Boyle (1996) stated that teachers should ask the following questions before using
authentic assessment in their music classrooms: Can we make time-consuming assessment
manageable? Can individual assessment time be managed? Is the assessment data reliable?
However, in spite of these questions about authentic assessment, the rich data gained through an
authentic process is valuable because of the difficulties inherent in assessing young children.
Welch (1994) pointed out that, when assessing musical behaviors (e.g., singing), the
nature of the sample, the location of assessment, the performance context, and the assessment
method are critical to the success of gathering data with young children. In discussing methods
for assessment of musical behavior in young children, McDonald and Simons (1988) asserted
that the assessment tool should:
1. be criterion-referenced,
2. include each area of music learning,
3. be referenced to specific instructional objectives,
4. be diagnostic, referenced to a child’s developmental stage,
5. measure behaviors that represent a child’s actual abilities,
6. be able to be scored consistently,
7. be conducted periodically so that it contributes to an overall evaluation of musical
growth, and
8. be administered easily (p.181).

	
  

9	
  

Qualitative and Quantitative Assessment
Since the publication of the Pillsbury Study (Moorhead & Pond, 1978), a seminal study
of children’s musical behavior, researchers have conducted a growing number of qualitative
studies of children’s musical behavior (Barrett, 2006; Berger & Cooper, 2003; Burton, 2002;
Hornbach, 2005; Kierstead, 2006; Metz, 1989; Moorhead & Pond, 1978; Reynolds, 1995;
Shelley, 1981; Young, 2002, 2003). They have chosen to work in the qualitative paradigm,
because qualitative methodology allows for the observation of children’s behavior in depth and
without disrupting the flow of their learning or their learning environment. While the qualitative
paradigm provides insightful perspectives of children’s musical behavior, there also is a need for
quantitative assessment, because quantitative measurement tools (e.g., checklist or rating scales)
are time efficient, easier, and quicker to score. Moreover, they allow teachers to assess a large
number of children, compare individual or groups, (Campbell & Scott-Kassner, 2006), and
gather group information about the measurement tool, such as reliability and validity. With
increasing demand for documentation for accountability and teacher evaluation, an efficient and
systematic way to keep track of children’s musical growth is critical. Yet few tools exist to help
teachers assess their young students’ musical growth.
Documenting children’s musical growth in a systematic and accurate manner so that
teachers know and have records of student’s musical learning, musical strengths, and musical
weaknesses is essential in order to meet children’s individual musical needs. “One size fits all”
music instruction that teaches to the average level of students can frustrate low
aptitude/achievement students and bore those students who are ready for additional challenges
(Taggart, 1989, 2005). Therefore, without proper assessment as a basis for differentiating
instruction, teachers will not have enough information upon which to base instruction so that it

	
  

10	
  

meets the needs of each student. As a result, some children will lose interest in learning music
and soon forget to “muse within” (Bjørkvold, 1992).
With this in mind, the development of a systematic measurement tool for use in assessing
and documenting children’s musical behaviors could be useful. In developing such a measure, it
is important that it can be used in a learning context without interrupting the flow of children’s
music activities and without children being conscious of being assessed; the assessment process
should be viewed as play for the children and still serve as an efficient tool for the teacher.
Purpose and Research Questions
With the intent of improving music instruction for preschool children and of providing a
measure that aligns with the National Music Content Standards (NAfME, 2012b), the purpose of
this study was to develop a measure to assess the musical behaviors of children in the context of
regular music instruction using continuous rating scales, to establish the validity and reliability of
the scales, and to develop appropriate assessment activities that can be used as part of instruction.
The specific research questions for the current study were as follows:
1. Can subjective validity be established for a researcher-designed instrument developed to
measure musical behaviors of preschool children in the context of music instruction?
2. What are the intrajudge and interjudge reliabilities of a researcher-designed instrument
developed to measure musical behaviors of preschool children in the context of music
instruction?
Delimitations
First, this study sought to describe and assess the musical behaviors of 4-, and 5-year-old
children who were enrolled at an inner-city private school in Detroit, Michigan. The students in
this school were mostly African-American, and 59 percent of the children in Pre-K program

	
  

11	
  

received reduced or free lunch. Therefore, the results of this study can be generalized only to
similar settings. Second, because this study sought to gain insight about children’s musical
behaviors in response to teacher-guided musical activities, the current music specialist, who also
was the researcher, taught all activities. An instrument is only valid when it corresponds with the
curriculum. In this school, Music Learning Theory (MLT) forms the foundation of the
curriculum. Therefore, the musical behaviors measured by the ECMBM were those that are
valued within this curricular approach. Finally, this study was limited to the investigation of
children’s musical behaviors exhibited in the context of music activities. Therefore, the data was
collected during the children’s regular music classes.

	
  

12	
  

CHAPTER II
REVIEW OF LITERATURE
Music Learning in Preschool Settings
Being with others is important for preschool children (Copple & Bredekamp, 2009), and
companionship and interaction among peers not only makes their musical experiences dynamic
but also helps them build their own musical culture or community (Campbell, 1998; Campbell &
Scott-Kassner, 2006; Flohr & Trollinger, 2010). Even though preschool is, by nature, a school
setting, children in these settings should be given ample time to play musically and be guided
informally. Unless children are given time to play and explore, school settings pose the danger
of forcing children into correctness rather than allowing them time to explore (Copple &
Bredekamp, 2009; Miller & Almon, 2009). Furthermore, because instruction that predisposes
the child toward learning includes motivational factors (Bruner, 1966) such as “play” (McDonald
& Simons, 1988), researchers have emphasized that music learning should be playful, child
centered, and teacher-facilitated (Andress, 1973; Campbell & Scott-Kassner, 2006; Taggart,
2000; Valerio, Reynolds, Bolton, Taggart, & Gordon, 1998).
NAfME (2012b) proposed content standards that state that prekindergarten children
should: (1) sing and play instruments, (2) create music, (3) respond to music, and (4) understand
music. As an extension of this, early childhood music curricula often include singing in a variety
of tonalities and meters, patterning (both tonally and rhythmically), moving to music, playing
simple percussion instruments (Valerio et al., 1998; Woods, 1989; Zemke, 2000), and acquiring
performance skills in beat keeping, pitch matching, and simple improvising (Woods, 1989).
A curriculum that is informal yet guided is ideal for young children. For example, formal
singing instruction - focusing on correcting pitches, learning the text, or singing based on adult-

	
  

13	
  

norms - is not developmentally appropriate, as children may not have developed their singing
voices for such formal instructions (Runfola & Rutkowski, 2010). As a result, rather than
teaching children to learn to perform specific songs, teachers should be acculturating children in
a variety of styles, tonalities, and meters while singing/chanting, moving, and patterning within
these contexts in order to build rich tonal and rhythm vocabularies (Gordon, 2013). All of this
eventually will be the basis of successful singing or chanting. This parallels the language
learning process, during which children are acculturated by listening to language as a basis for
successful speaking. Therefore, the main focus of early childhood music instruction should be
on designing curricula that guide children to acculturate and understand music within a context
(i.e., tonality and meter), and build musical vocabularies to help children to acquire readiness for
success in their formal music instruction later on.
Musical Behaviors of Preschool Children
Musical behaviors are what children do musically either intuitively or as a result of
earlier experiences or instruction, and they usually occur in relation to a musical context, person,
or stimulus (Berger & Cooper, 2003; McDonald & Simons, 1988; Moog, 1976). Of various
musical behaviors, researchers have identified singing, chanting, moving to music, and playing
instruments as the prevalent musical behaviors that have been observed in preschool children
(Andress, 1973; Aronoff, 1969; Campbell & Scott-Kassner, 2006; McDonald & Simons, 1988;
Metz, 1989; Moog, 1976; Moorhead & Pond, 1978; Welch, 1998; Zemke, 2000).
In describing the musical behaviors of children, understanding the developmental stages
is important, as “considerable growth and change occur in children during the preschool years in
all areas of development” (Copple & Bredekamp, 2009, p.112). Developmental learning theories,
such as Piaget’s (1950) “stages of cognitive development” or Bruner’s (1966) “theory of

	
  

14	
  

instruction,” propose a set of developmental stages that children go through sequentially and
provide windows for music educators to gain insights into the development of children.
One of the most extensively used developmental theories in music education that
describes children’s musical behaviors is Bruner’s “theory of instruction” (Aronoff, 1969;
Boardman & Andress, 1981; Campbell & Scott-Kassner, 2006; Flohr & Trollinger, 2010;
McDonald & Simons, 1988). In an application of developmental learning theory in the music
domain, Aronoff (1969) and Boardman and Andress (1981) have described musical behaviors
according to Bruner’s (1966) three levels of intellectual functioning that children display:
enactive representation, iconic representation, and symbolic representation. Preschool children
are in enactive representation when they demonstrate musical understanding through physical
movement, imitation, and show their musical ideas by manipulating instruments (Boardman &
Andress, 1981); they are in iconic representation when they transform musical experiences into
aural, kinesthetic, and visual images (Aronoff, 1969). Through iconic representation, children
may recall rhythm patterns (aural), repeat movement - galloping, walking, running – with music
that represents these movements (kinesthetic), or improvise new rhythmic patterns (Aronoff,
1969). Visually, iconic thinking could be some kind of “pictorial doodling” to represent music
that children may further develop into “invented notation” (Barrett, 1997) and, later, into
traditional notation, which then is displayed through symbolic representation. Symbolic
representation is displayed through use of words or symbols. Children perform, describe
verbally, read, and decode their own invented notation and/or traditional notation through
symbolic representation.
In language, reading and writing in early childhood is exploratory in nature; it is not
considered developmentally appropriate to teach children to read the actual words or sentences,

	
  

15	
  

but it is appropriate for children to “pretend” to be reading or writing (Miller and Almon, 2009).
Children may learn to read and write on their own through this process, but reading and writing
is not taught directly at first. In music, while children can explore invented and real notation
early on, it should not be the focus of learning or be imposed during early childhood music
instruction, because the process tends to be task-specific (Barrett, 2006; Welch, 1998); also,
there is an end product, which conflicts with the nature of informal instruction. Similarly,
although “pictorial doodling” of music often is spontaneously displayed by children, learning
traditional notation through symbolic representation should be avoided in early childhood, as
children are not developmentally ready for such music literacy, which requires formal instruction.
In addition to developmental factors, external factors such as experience, education,
family, peers, and socio-cultural context affect the musical behaviors of growing children (Welch,
1998). Welch explained:
As the child grows and develops, the actual musical behaviours and capabilities that
emerge are the product of a complex interaction between, on the one hand, general
intellectual predisposition and potential, the species-wide capacity for musical behaviours,
and on the other, particular environmental experiences that, to a greater or lesser extent,
‘match’ and allow such potential to be realised (p. 29).
The context in which the behavior occurs also is important to consider when describing
children’s musical behavior. Zimmerman (1981) remarked, “Our understanding of musical
behavior is only as good as the context in which that behavior is observed” (p. 39; as cited in
Veldhuis, 1992, p. 6), and McDonald & Simons (1988) also indicated that different situations
tend to elicit different musical behaviors.

	
  

16	
  

Another factor that affects the eliciting of musical behavior in the context of a music
class is the musical “others” in the setting. Sullivan (2006) identified two types of musical
behaviors in the context of music class: independent behavior – behavior exhibited without
influence of other/s, and dependent behavior – behavior exhibited directly from what others do,
usually a response to or imitation of others. When children sing or chant spontaneously,
children’s musical behaviors lean more toward independent behavior. Children singing a song
while playing by themselves with manipulatives or humming songs as they walk to their
classrooms are self-elicited, because the child is the initiating force and the environment is
secondary. Examples of dependent behavior are pattern echoing, imitating the teacher’s
movement, or improvising rhythm patterns after teacher-initiated patterns. Upon entering
preschool, dependent musical behavior is observed frequently, as children encounter the “social
culture” of schooling, experiencing the culture as a member of the group.
Singing Behavior
The most common tonal behavior children exhibit throughout their musical development
is singing. Welch (1994) remarked, “Singing is a highly complex human behaviour which is
developmental in nature and which exists in many different forms amongst the musics of the
world” (p. 16). The fetus begins to develop the ability to hear in the second trimester of
pregnancy (Gembris, 2002) and accumulates the urge to muse-within (Bjørkvold, 1992) through
internal sound: the mother’s voice streaming down on the spinal cord, heartbeat, or movement of
internal organs (Lecanuet, 1996). With their first outcry, newborns finally release the “urge” and
signal their initial ability to vocalize and later, sing.
Children’s vocal ability develops sequentially (Welch, 1998). Newborns first begin to
coo and babble in response to musical stimuli and engage in vocal-play with their parent. By age

	
  

17	
  

two, many children match pitch and imitate parts of heard songs. When children reach ages four
or five, some can sing entire songs with somewhat precise melodic and rhythmic contour, though
accurate pitches typically have yet to be mastered (Sloboda, 1985).
Several researchers have noted that certain singing behaviors have primacy over others
according to that developmental sequence (Moog, 1976; Sloboda, 1985; Welch, 1994; Welch,
Sergeant, & White, 1996). After moving beyond the babbling phase, children cease to babble
and begin to focus on fragments of song text followed by rhythmic features, pitch components,
and key stability (Welch, 1994). Similarly, Gordon (2013) remarked that, when a child is ready
to sing or chant, the child exits from rhythm babble or tonal babble.
Singing is a central component of music education for children in the Western world, and,
as with any other musical behaviors, singing skill development is both context and person
dependent. While a child may sing spontaneously with utmost freedom while at play, that same
child may adjust his or her singing behavior when singing in a group or for the teacher in order
to meet established musical expectations, such as when making an effort to sing in tune to match
the pitch of others (Andress, 1980). Although the teacher may not ask for these singing
adjustments, as giving formal instruction is not developmentally appropriate with preschool
children, exposure to cultural context results in children changing their musical behaviors as a
result of context.
Movement Behavior
Movement has been recognized as one of the first natural responses that children produce
musically (Gordon, 2013; Hicks, 1993; Metz, 1989; Moog, 1976; Moorhead & Pond, 1978;
Reynolds, 1995). As a growing child develops, his or her movement range increases in terms of
responding to and expressing music. Eventually, children learn to move to music with intention

	
  

18	
  

and control. Moog (1976) investigated the musical behaviors of 500 children to musical stimuli
in their homes and observed that toddlers began to move intentionally, which contrasted with the
spontaneous moving without intent that they did in their earlier years. He also observed that,
whereas infants naturally respond to musical stimuli with whole body movement, as children get
older, they tend to use individual body parts in their movements and learn to coordinate their
movement to music.
In their first year, children can acquire a movement repertoire through informal musical
experiences, during which they mainly watch others and acculturate to their musical
environments (Gordon, 2013). Then, children begin to use their movement capabilities - walking,
stepping, swaying side to side, twirling, or waving arms – to engage in creative movement,
though the movement is not rhythmically coordinating to the beat of music (Edwards, 2013;
McDonald & Simons, 1989; Sloboda, 1985). With physical and musical maturation, preschool
children initiate structured movement, such as clapping, marching, dancing, or playing sticks,
and work to synchronize it to music. By the time children reach age five, most have acquired a
variety of movements, and their spontaneous movements to music seem to decrease (Moog,
1976). Also, their movement is likely to be elicited upon request or when movement is visually
modeled for them (Metz, 1989). It can be speculated that the reason behind the decrease of
spontaneous movement during the preschool years is that children’s increased ability to imitate
(McDonald & Simons, 1988; Sloboda, 1985) and their language skills provide other
opportunities for expressing themselves.
In the context of music class, children acquire movement repertoire through teacher
demonstration, modeling, and peer imitation (Campbell & Scott-Kassner, 2006). Particularly,
the presence of a guiding adult plays an important role in building movement repertoire, as it

	
  

19	
  

encourages children’s movements to become more imitative and interactional. Metz (1989)
investigated the movement responses of preschool children at a researcher-designed music center
in which children choose to participate during free-play time. She found that the most movement
responses resulted from “child-adult” interaction in which the researcher modeled, described,
and suggested various movement responses for and to the child.
The properties of elicited movements (i.e., modeling, describing, and suggesting)
proposed by Metz (1989) are incorporated frequently during music activities as teaching
strategies. In addition to modeling movements for children, teachers often suggest movements
verbally, such as “Can we move like?” or by asking questions like, “How should we move?” to
encourage movement from children who are either reluctant to move or not used to moving their
bodies to music. These strategies are considered to be developmentally appropriate, as teachers
do not impose movements on children but guide them to become aware of moving their bodies
(Edwards, 2013).
In an investigation of creative movements of 3-, 4- and 5-year-old children, Sims (1985)
developed a Movement Observation Form (MOF) and recorded categories of movements based
on four types of movement categories: locomotor, axial, small motor, and no movement. Her
study revealed that, of 3- to 5-year-old children, 3-year-olds showed the most “no movements”
(50%), followed by small motor, locomotor, and axial movement, respectively; 4-year-olds
showed a high percentage of axial (33.59%), followed by small motor movement (32.81%), no
movement, and locomotor; and 5-year-olds showed a high percentage of locomotor movements
(61.61%), followed by axial, no movement, and small motor movement. For the same study,
she developed a Rhythmic Movement Observation Form (RMOF) to investigate the rhythmic
characteristics of preschool children. She classified three rhythmic movement types in the

	
  

20	
  

RMOF: (1) moves rhythmically 90-100% of the time but not necessarily synchronizing with the
beat of the music, (2) movement corresponds to the beat at least 90% of the time, and (3) non1

rhythmic. The results indicated that percentage of “rhythmic movement” increased with age:
the movement of 3-year-olds was 22.49% rhythmic, the movement of 4-year-olds was 61.97%
rhythmic, and the movement of 5-year-olds was 73.86% rhythmic. Sims (1985) concluded that
task and degree of body control are important factors along with age and experience in
expressing rhythm through movement.
Rhythmic Behavior
According to Gordon (2012), there are two main components of rhythm in music: beat
(macrobeat/microbeat, which comprise meter) and melodic rhythm. Children’s inherent ability
to move to music is in response to these rhythmic components in music, though their movement
may not express the rhythm patterns or be synchronized to beat. However, children increase
their abilities to coordinate movement in relation to rhythm, particularly to beat, during the
preschool years, as these skills are acquired through experience and increase with age (Metz,
1989; Moog, 1976).
Moving to beat is perceived as an integral component of preschool children’s music
learning, and, because it demonstrates children’s perception of rhythmic abilities, researchers
have designed performance tasks to measure children’s rhythmic behavior or abilities through
various modes. Mainly, they measured beat keeping (1) using small muscles (clapping or
tapping) (2) with instruments or (3) by moving to music in a locomotor way, which often is
displayed during performance through dance (Malbrán, 2002; Rainbow, 1981; Rainbow & Owen,
1979; Schleuter & Schleuter, 1985).
	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
1
A term “operationally defined in this [Sims’] study as a series of at least three repeated
movements or movement patterns which were synchronized to a steady beat” (Sims, 1985, p.44).
	
  

21	
  

Rainbow and Owen (1979) conducted a 3-year longitudinal study to investigate the
rhythmic abilities of 3-, and 4 –year-old children. The results of this longitudinal study revealed
that, of the following rhythm tasks: (1) motor tasks using small or large muscles (keeping steady
beat using hands or feet or a combination of both); (2) motor tasks using simple instruments
(keeping steady beat using rhythm sticks); and (3) vocally producing rhythm pattern (speaking
rhythm pattern then clap or speaking rhythm ostinato to music), performing rhythm vocally was
the least difficult, followed by keeping beat by clapping and with rhythm sticks. Marching to
music, marching and clapping rhythm and reading ostinato to music were difficult for these
children. In the final report of this longitudinal study, Rainbow (1981) reported that, when given
14 rhythmic tasks - five tasks of which related to performing steady beat to recorded music and
nine tasks of which related to performing rhythmic patterns that children echoed either vocally or
by clapping - children’s vocal response was more successful than their movement response.
However, the “vocal mode” was used for the rhythmic pattern tasks and not for the tasks related
to beat.
Schleuter and Schleuter (1985) found a similar result with Kindergarten through third
grade children. After one year of music instruction of clapping, chanting, and stepping sound
patterns, chanting resulted in the most accurate responses for Kindergarten students and students
in grade 1, and clapping resulted in the most accuracy for students in grades 2 and 3. The
rhythm task in this study was performing rhythm patterns rather than keeping beat. From the
studies on children’s rhythm abilities (Rainbow & Owen, 1979; Rainbow, 1981; Schletuer &
Schleuter, 1985), it can be speculated that preschool children’s performances of rhythm patterns
are most successful vocally, but keeping beat can be achieved most accurately through tapping or
playing simple percussion instruments.

	
  

22	
  

Chant often is associated most with rhythm patterns, is not pitched specifically, is more in
speaking voice than singing voice, and is produced vocally. In the Pillsbury Studies, Moorhead
and Pond (1978) described chant as a type of rhythmic song that was “sung most often in the
group, usually loudly, repeated over and over again, rising often to a high emotional pitch” (p. 8).
For example, a single child might start a chant, and other children would join in, noticing the
“play component” of the chant; the children identify a musical fragment to repeat over and over
again with one another for fun. Unlike spontaneous singing that typically is produced in solitude,
chanting is exhibited predominantly in social settings and often is performed in conjunction with
large movement activities or materials (i.e., props or percussion instruments). In a structured
context, children’s chant behaviors often are initiated and guided by the teacher with variation in
dynamic, tempo, or vocal contour.
Instrument Playing
During the early childhood years, children’s musical instrument playing behaviors are
usually in the form of free exploration or accompanying singing or recorded music (Miller, 1986;
Moorhead & Pond, 1978). When instruments are presented to children for the first time, they
explore their sound capabilities rather than using them to produce sound that is related to tonal or
rhythmic elements of music (McDonald & Simons, 1988). In the exploratory stage, children
play with the sound for its own sake (Flohr, 1984; Moorhead & Pond, 1978; Young, 2003).
However, they soon begin to coordinate their movements to play the instruments to a steady beat
or to produce rhythm patterns (McDonald & Simons, 1988). With the increase in the
coordination of their gross motor skills as they grow older, children begin to develop the skills to
play both rhythmic and pitched instruments within a musical structure, incorporating the various
musical elements such as patterns, tempo, or meter.

	
  

23	
  

Creativity and Improvisation
“Creative musical behaviors manifest themselves long before children can read or write
music” (McDonald & Simon, 1988, p.49). Musical improvisation is a form of children’s
musicing. However, only with abundant exposure to and opportunities to engage in music play
or movement play will children acquire the music vocabulary to imitate and extend to creativity
and improvisation (Valerio et al., 1998).
Whitcomb (2010) found that preschool children are able to improvise both simple and
complex rhythm patterns with an underlying steady beat on drums. She investigated preschool
children’s (N = 6) rhythmic improvisations on drum, and, after 5 weeks of their participating in
“call-and-response” activities, found that more than 79% of the time children were able to
improvise both simple and complex rhythmic patterns within the following specific guidelines:
start immediately following the call, end on last beat, improvise for four beats long, and
improvise containing a steady beat.
In comparison to possible structured improvisation on percussion instruments in the study
of Whitcomb (2010), improvising tonally on a pitched instrument, such as Orff xylophone tends
to be more exploratory in nature (Flohr, 1984; Young, 2003). Children’s striking and playing
glissandos represented the few improvisatory musical behaviors in which children engaged on
xylophones when observed in a natural environment (Young, 2003). Flohr (1984) investigated
the improvisatory behaviors of 2-, 3-, and 4-year-old children on xylophone longitudinally and
found that the characteristics of the children’s improvisations changed in relation to their
chronological age. Two-year-old children were interested in exploring sound and timbre,
whereas 3-year-old children often imitated in response to researchers’ phrases. Four-year-old
children repeated pitches and patterns most often, although they were not within a structured

	
  

24	
  

tonal system. However, 4-year-old children were able to structure the rhythmic elements of their
improvisations.
Assessing Musical Behavior of Preschool Children
Musical behavior is multidimensional, and the dimensions (i.e., singing, chanting,
moving, improvising, playing instrument) often occur in a synthesized form, which makes
assessing musical behavior difficult. Therefore, when assessing children’s musical behavior, it is
necessary to focus on a single dimension at a time in order for assessment to result in an accurate
and comprehensive picture of children’s multidimensional abilities in music (Gordon, 2002).
Although assessing children’s musical behaviors in an individual test setting might be
possible for teachers, behavioral observations in familiar settings might be more effective and
more developmentally appropriate for children, because many preschool children feel distress
and discomfort in a non-naturalistic testing setting (Greenberg, 1976). As a result, they are not
able to display their musical abilities fully. Young children also are reluctant to perform on
demand. Therefore, instead of assessing children in an individual testing setting, it may be more
appropriate to observe their “natural forms of musical expression” (Moorhead & Pond, 1978),
which is best displayed by their behaviors in a familiar context.
Theoretical Frames
Theoretical frames that are useful for observing children’s musical behavior from the
perspective of musical development (Gordon, 2013; Welch et al., 1996) are important in
designing developmentally appropriate assessment. Early childhood music programs typically
function within a theoretical frame that should play a role in determining how to assess
children’s musical behaviors and which behaviors to assess. The theoretical frame provides a
context for observing children’s behaviors and guides teachers in knowing where children are in

	
  

25	
  

terms of their music development in relation to the framework. Music Learning Theory (MLT),
one such framework, provides a clear outline of how children move through music development.
Gordon (2013) noted that, as children continue to interact musically with those around them,
they progress through theoretical types and stages of music development. He proposed three
types and seven stages of preparatory audiation. When observing children’s musical behaviors,
a teacher can place the observed behaviors within those types and stages, allowing the teacher to
assess children’s behaviors sequentially within the theoretical frame. Early childhood music
programs based on MLT uses the stages of preparatory audiation as a framework for
understanding and assessing children’s musical behaviors (Etopio & Cissoko, 2005; Taggart,
Alvarez, & Schubert, 2011; Valerio et al., 1998).
Based on research on children’s singing, Welch et al. (1996) proposed a model of Vocal
Pitch-Matching Development that can serve as a frame when observing children sing songs. The
model included four sequential stages. Following stages are that describe children’s vocal
development: Stage 1 - the words of the songs are the main interest rather than melody, and
singing is like chant; Stage 2 - children become aware of vocal pitch; Stage 3 - children’s
melodic shape and pitches are mostly accurate, although some tonality changes may occur Stage 4, no significant melodic or pitch errors occur.
Measurement Tools
NAfME (2012a) suggested the following methods for assessing young children: (1)
checklists or anecdotal reports completed by teachers, parents, or aides to record and describe
verbal and nonverbal behaviors; (2) systematic observation documenting such behavior as time
on task, number of instances of an event or behavior, and participation tendencies over time; and
(3) rating scales to provide data related to quality of responses, such as degrees of accuracy,

	
  

26	
  

originality, or involvement. Anecdotal notes or measurement tools such as rating scales were
used in research studies investigating children’s musical behavior (Kierstead, 2006; Metz, 1989;
Miller, 1986; Moorhead & Pond, 1978) or to measure specific musical abilities or achievements
(Blesedell, 1991; Levinowitz, 1989; Rainbow, 1981; Rainbow & Owen, 1979; Rutkowski, 1990;
Taggart, 1994).
Checklists or Additive Rating Scales
Rating scales can be used to assess the quality of child’s performances and degrees of
competence in many developmental areas (Bentzen, 2005). Researcher- or teacher-designed
measurement tools pertaining to the observation of children’s musical behaviors are most often
checklists or additive rating scales identifying the presence or absence of a behavior or the
frequency of occurrences of that behavior (Fox, 1989; McCusker, 2001; Miller, 1986, Shelley,
1981). In general, a checklist can be used to record occurrences of specific behavior and
specifically whether a child can achieve a task (Bentzen, 2005).
The Music Study Observation Form reported in Shelley’s (1981) study was a taxonomy
of 19 musical behaviors and was constructed of checklists to identify frequencies of those
behaviors during free activity time within a preschool class at the University of Maryland’s
Center for Young Children (CYC). Behavior descriptions, such as “observes but does not
participate” or “examines and manipulates instruments” were listed, and the researchers recorded
occurrences of those specific behaviors. The reliability of the measure was not provided in the
study.
Miller (1986) constructed the Musical Behavior Observation Matrix (MBOM) to assess
children’s spontaneous musical and social behaviors in a naturalistic setting. When constructing
the observational checklist, Miller (1986) first observed children’s musical behaviors in a

	
  

27	
  

naturalistic context, such as free play at a researcher-designed music center, and used event
sampling to identify and categorize behaviors. During phase one, children’s behaviors were
grounded (Strauss & Corbin, 1990); that is, she identified and determined musical behaviors that
children commonly displayed during a 30-minute observation. Based on the CYC study, she
noted the frequency of occurrence for each behavior. The MBOM included 30 of the most
common behaviors of children as identified in her earlier research. Independent judges observed
the videotape and used event sampling to record behavior. The inter-observer reliability was .92.
McCusker (2001) observed children’s musical behavior in an early childhood music
program and developed an assessment inventory to record the range of children’s musical
behaviors over time. The measurement tool is a checklist through which children’s musical
behaviors in singing, chanting, moving, playing, literacy, and literacy-related events can be
assessed and recorded in a classroom context. Her inventory form is designed to allow teachers
to compare one child’s musical behaviors to those of other children and to assess each child’s
musical development over time. Because the musical inventory list for children’s musical
behaviors was constructed for qualitative analysis in order to understand children’s emergent
musical literacy, reliability was not reported.
Numerical Rating Scales or Rubrics
The rating scale has many advantages as a tool in observation, as it is less time
consuming to use, convenient for observing a large number of traits, and easy to score and
quantify than most other methods. MusicTIME Behavior Profile (Fox, 1989) was intended for
parents whose children were enrolled in the early childhood music program at the Community
Education Division of the Eastman School of Music. The measure was designed for use at home
rather than in the context of an early childhood music class as a way for parents to communicate

	
  

28	
  

with music teachers about their children’s musical behaviors and as a way to help parents
become more aware of the musicing of their children. Parents were given the profile to take
home before the last class and were asked to rate their children’s musical behavior. The profile
consists of 55 items concerning musical behavior, grouped into the categories of vocal
development, movement response, exploring sounds, and participation. Each individual
behavior is coded according to its frequency or consistency of occurrence at home. MusicTIME
Behavior Profile is more specific than a checklist, as it provides numerical measure indicating
frequency of occurrences. The reliability of this measure was not reported. Parents are asked to
give scores of 0 = not observed, 1 = seldom observed, 2 = frequently observed, and 3 = always
consistently observed. In addition to numerical scoring, parents also are asked to comment on
their ratings. The purpose of the assessment is to keep track of children’s musical growth
longitudinally. Although the descriptions provide detailed information about the presence and
frequency of the child’s musical behaviors, the given numerical rating does not inform the
teachers or parents where the children are in terms of their musical development; that is, because
the behaviors are not contextualized within a developmental theoretical framework, it is difficult
for the parents to understand where their child is in terms of musical developmental processes.
In order for teachers and parents to evaluate children’s musical development, a theoretical frame
that explains the child’s development on a continuum would be useful.
The Test of Early Audiation Achievement (T-EAA) was developed by Runfola and Etopio
(2006) over a 3-year period as part of a project funded by the National Endowment of Arts
(NEA) (as cited in Lee, 2010) to measure the tonal audiation and rhythm audiation achievements
of Pre-K children (Etopio, 2009). T-EAA has subtests: tonal center (resting tone), tonal patterns,
rhythm patterns, and singing a familiar song with tonal and rhythm dimensions. Children are

	
  

29	
  

assessed based on rubric with scale of 0-4, and this particular measure (rubric) is designed for
individual testing rather than for use in a class setting. The content validity was established by
constructing items that were consistent with the first two types of preparatory audiation as
defined by Gordon (2013) and his hierarchy of tonal pattern and rhythm pattern difficulty levels
(Gordon, 1976). The assessment requires children to perform multiple items for each tonal
(tonal center, tonal pattern, singing a familiar song: tonal) and rhythm subtest (rhythm pattern,
singing a familiar song: rhythm). The interjudge reliability of the measure, among subtests was
moderate to high (tonal center = .88; tonal patterns = .92; rhythm patterns = .91; singing a song:
tonal = .73, rhythm = .63).
Continuous Rating Scales
Continuous rating scales are useful, because they indicate where children are on a
musical continuum and give diagnostic information to teachers for use in guiding their
instruction. Continuous rating scales include criteria, and reliabilities tend to be higher on
continuous rating scales when criteria are specific and sequential. However, although the
specific behavioral descriptions within a developmental context contribute to reliability and
validity of the measure (Gordon, 2002), some rating scales are too detailed and too long, making
them difficult to use in a class context. Therefore, the description of behaviors within the criteria
should be succinct, yet specific, and appropriate for use in a formal assessment.
The investigation of studies using continuous rating scales revealed that, unlike additive
or numerical rating scales, continuous rating scales can be used successfully to measure specific
dimensions of musical behavior, ability, or achievement (Hornbach & Taggart, 2005; Levinowitz,
1989). In the following section, I will discuss the continuous rating scales that have been used to

	
  

30	
  

measure following dimensions of children’s musical behavior: singing, pattern performance,
movement, and rhythm, as well as the use of these scales in the context of music instruction.
Singing. Most studies measuring singing behavior suggested that children be assessed
individually in one-on-one setting (Etopio, 2009; Rutkowski, 1990, 1996; Sundin, 1997). For the
rating of song performance, researchers either include dimensions of pitch and rhythm in their
rating scale (Levinowitz, 1989) or assess singing holistically (Rutkowski, 1990; Sundin, 1997).
Rutkowski’s (1990) Singing Voice Development Measure (SVDM) was developed “to
establish a more consistent means of describing the various stages of child singing voice
development and to provide a consistent means for teachers and researchers to more accurately
measure and describe the use a child has of his/her singing voice” (Rutkowski, 1997, p.202; as
cited in Rutkowski, 2010). “SVDM has been developed over a period of 25 years to measure the
use a child has of his or her singing voice; it is not measurement of accuracy of intonation”
(Rutkowski, 2010, p. 273). The current measurement is 5-point continuous rating scale with
subdivisions of the five points into a total of nine levels. However, since nine levels would be
quite difficult to use, five scores are used as anchors (Rutkowski, 2010).
Studies assessing children’s singing behavior and other studies using Rutkowski’s rating
scale have indicated that the rating scale is valid for use with children from 3 years old up to
those in grade 3. Studies have used SVDM with kindergarten and elementary children by giving
them the specific singing task of echoing patterns of the criterion song (Rutkowski, 1990, 1996;
Rutkowski & Miller, 2003), rather than observing their naturalistic singing behavior in the
context of a class setting. In Rutkowski’s studies, the validity was based on interjudge reliability,
which ranged from .80 - .99. She also established subjective validity by consulting with child
voice experts when developing the scale.

	
  

31	
  

Although SVDM is used widely and is the closest to a standardized measurement tool for
singing voice, it may be difficult for a teacher to use during activities in a classroom because the
rating has too many scale items (nine subdivisions). Therefore, Rutkowski (2010) recommended
that, for informal assessment, a few children during each class should be targeted during music
activities, and, for formal assessment, children should echo patterns and audio-record the activity
so that teacher can assess after class. If it were to be used in a group setting, teachers would need
to design assessment procedures carefully, because the procedures are time consuming and
children would need to wait quietly while their peers were being assessed individually.
Levinowitz (1989) constructed a 5-point continuous rating scale with tonal achievement
and rhythm achievement dimensions to assess children’s singing with and without text. She
reported high interjudge reliabilities for both the tonal and rhythm dimensions. The interjudge
reliabilities for tonal achievement were as follows: songs without words tonal, .78; songs with
words tonal, .93; tonal composite, .87; songs without words rhythm, .84; songs with words
rhythm, .94; and rhythm composite, .90. For both the rhythm and the tonal rating scales, the
interjudge reliabilities were higher for the songs with words than for the songs without words.
Although high reliabilities were reported, the detailed description for each criterion in the tonal
achievement rating scale would make it difficult to use in a class context. In the tonal dimension,
the judge needs to identify tonic or dominant patterns in the criterion song and be able to rate
based on the functions of each pattern. This makes the rating scale less useful for a wide range
of repertoire, as different repertoire would have different tonal functions.
Pattern Performance. Taggart (1994), in her validity study of Audie (Gordon, 1989), a
test of music aptitude for 3- and 4-year-old children, constructed 5-point rating scale for use in
rating tonal and rhythm pattern performance in the context of regular music activities. Easy,

	
  

32	
  

medium, and difficult patterns were given to children during music activities by their music
teacher, and independent judges rated the children’s performances. The interjudge reliabilities of
tonal dimension were as follows: major tonality, .98; minor tonality, .86; and composite .96.
Interjudge reliabilities of rhythm dimension were duple meter .99; triple meter, .93; and
composite .98. These reliabilities are high and provide evidence of the scale’s validity.
However, to get this level of reliability, it required that two judges sit in the classroom while the
teacher was fully engaged in teaching, so this is less practical for use by a classroom teacher.
Movement and Rhythm. In order to investigate the effect of movement (Dalcroze or
Laban method) instruction on the rhythm achievement of 3- and 4-year-old children, Blesedell
(1991) constructed a movement performance rating scale and a rhythm performance rating scale
to rate children’s performances. Independent judges rated children’s performances, and the
interjudge reliabilities for the movement rating scale were .51, .55, and .73 and were .59, .70,
and .73 for the rhythm rating scale. Children’s movement and rhythm performances were
videotaped for rating during several activities in the first, fifth, and tenth lessons. The
performance task for rhythm achievement included a criterion song, to which children performed
micro- or macrobeats on a small drum. For the movement task, children who had been exposed
to Laban-based instruction participated in the execution of movement story written by the
researcher. All lessons focused on four of the movement themes described by Laban: (1) body
awareness, (2) awareness of weight and time, (3) awareness of space, and (4) awareness of flow.
Children who participated in Dalcroze-based movement lessons were exposed to body
movement activities adapted from Findlay’s (1971) application of Dalcroze eurhythmics. Three
independent judges watched the videotapes of activities for the movement assessment. The

	
  

33	
  

reliabilities for movement scale were low (Judge 2 to Judge 3 was .55; and Judge 1 to Judge 3
was .51) and calls into question the validity of the movement rating scale.

	
  

34	
  

CHAPTER III
METHODOLOGY
Subjects
The subjects in this study were 4- and 5-year-old children enrolled in a Pre-K program at
an inner-city private school located in Detroit, Michigan. This school accepts children ages 3
and 4 into the Pre-K program. Although the subjects’ ages were 3 and 4 at the beginning of the
school year, many had turned 4 and 5 by the time of data collection. Most of the 3- or 4-year-old
children were attending school for the first time. Children who were enrolled at this particular
school were mostly African-American, and 59 percent of the children in Pre-K program received
reduced or free lunch. The school had a total of six Pre-K classes, two at the main campus and
four at satellite campuses. Children (N = 36) in two intact classes participated in the main study.
They had music class for 40 minutes, two times per week.
The music curriculum at this school was based on Music Learning Theory (MLT), and
activities were designed to provide informal and structured guidance in music. The teacher used
the resources Music Play (Valerio et al., 1998) and Jump Right In: Kindergarten (Taggart,
Reynolds, Valerio, Bailey, Lange, & Gordon, 2009) in constructing music activities. Typical
music instruction included singing and chanting with text and without text in a variety of
tonalities and meters, tonal and rhythm pattern instruction, movement activities, rhythm
improvisation, and instrument playing. Children played simple instruments such as egg shakers,
rhythm sticks, or maracas in conjunction with singing, and played drums when keeping beat to a
song. Children also were guided to move using elements of movement: flow, weight, space, and
time (Laban, 1971) in a sustained and continuous way and with pulsation (Valerio et al., 1998).
During pattern instruction, children imitated tonal and rhythm patterns and improvised teacher-

	
  

35	
  

initiated rhythm patterns. All activities were provided in a playful manner, and no correct
responses were required of children.
Development of the Musical Behavior Measure
While observational assessment in a naturalistic context such as free play is
developmentally appropriate and is used in early childhood classrooms, this type of naturalistic
observation can be difficult in a music class, as music activities are ongoing, and teachers are
singing, chanting, and moving constantly to guide music activities. In addition, “free musical
play” often is not included in the curriculum, because music teachers only see children one or
two times per week and want to use their brief time with the students for structured guidance.
Therefore, in order to assess children’s musical behavior efficiently and keep track of children’s
progress, there is a need to develop a measure that can be used in the context of regular music
activities. This aligns with DAP, which recommends that the assessment process be woven into
teaching and interacting with children and incorporated in an on-going way by the teacher while
teaching.
Musical behavior in this study refers to the child’s intentional musical behavior generated
in response to the teacher, which is what typically occurs in a music class with children at this
age. Assessing this type of musical behaviors in the context of music activities is necessary to
assist a teacher in gathering information on children’s current musical abilities and in planning
future music activities.
Characteristics of a Musical Behavior Measure
The Early Childhood Musical Behavior Measure (ECMBM) was designed as an
authentic assessment, because it was intended for use with children who are engaging in typical
activities in a music class. Because “the choice of using a given type and design of a rating scale

	
  

36	
  

is decided best by those who are teaching or who have taught the students with whom the rating
scale is to be used” (Gordon, 2002, p.15), I, as the children’s music teacher, designed the scales
based upon my knowledge of the children and their levels of music development. In addition,
my work was informed by the work of past researchers and teachers who have developed music
measurements for young children. I chose to construct continuous rating scales, as they
contribute to reliability and can serve as a guide to the judges by providing specific descriptions
of children’s musical behavior in a developmental context. With this information, a teacher can
adapt instruction to meet children’s individual needs, which is the primary purpose of
measurement. In the measure, I explained unfamiliar vocabulary and provided additional notes
at the bottom of the rating scale sheets for the judges, as doing so contributes to higher reliability
and validity (Gordon, 2002).
Development Procedures
Development procedures for ECMBM were as follows:
1. Construction of dimensions and criteria for the rating scale: The researcher constructed a
preliminary design for ECMBM. Research literature, other measures of musical
ability/behavior, and Pre-K content standards proposed by NAfME (2012b) provided the
basis for the content.
2. Construction of Assessment Activities: I designed assessment activities based on the
content of the measure, the music curriculum of the school, Pre-K achievement standards
proposed by NAfME (2012b), and the review of literature on Pre-K music curricula and
assessment.
3. Preliminary Validation: The preliminary ECMBM and assessment activities were
presented for feedback and comment to the researcher’s advisor. She responded to

	
  

37	
  

ECMBM’s dimensions, criteria, and the connections between the assessment activities
and the measure. At this time, I was conducting a pilot study on the use of the measure
with children.
4. First Revision: ECMBM was revised based on my advisor’s comments and feedback and
what I learned during the rating process of the pilot study.
5. Validation Process: The revised version of ECMBM and assessment activities were
presented for feedback to a group of seven experts (university professors) in the field of
early childhood music education and measurement.
6. Final revision: The final revision was made to ECMBM based upon the feedback from
these experts.
7. Verification: Two validation panel members from the validation panel and the
researcher’s advisor verified the final version of ECMBM for its use in the main study.
Construction of Dimensions and Criteria
ECMBM was designed to gather information on the types of musical behaviors preschool
children display in response to music or in music-oriented activities in standard research studies
and in most early childhood music curricula. The Pre-K content standards of NAfME (2012b)
propose the following: (1) singing and playing instruments, (2) creating music, (3) responding to
music, and (4) understanding music. In this study, the following dimensions were determined to
be appropriate for assessing children’s musical behavior in the context of music activities: (1)
Tonal Behavior: Singing Accuracy, (2) Tonal Behavior: Resting Tone, (3) Tonal Behavior:
Tonal Pattern Imitation (used with patterns both in major and minor tonalities), (4) Rhythm
Behavior: Chant Performance, (5) Rhythm Behavior: Rhythm Pattern Imitation (used with
patterns both in duple and triple meters), (6) Rhythm Behavior: Rhythm Improvisation, (7)

	
  

38	
  

Rhythm Behavior: Keeping Beat, and (8) Movement Behavior. The rating scale dimensions for
2

each of those types of musical behaviors can be seen in Figure 1.
Figure 1
The Early Childhood Musical Behavior Measure

TONAL BEHAVIOR: SINGING ACCURACY
5. The child sings the entire song with correct melodic contour and on the correct pitches
with good intonation.
4.	
  The child sings the song with correct melodic contour but with a few incorrect pitches
or poor intonation.
3. The child sings the song with correct melodic contour but on the wrong pitches.
2. The child sings the song with some correct melodic contour but on the wrong pitches.
1. The child sings with a lack of tonal context and melodic contour is absent.

TONAL BEHAVIOR: RESTING TONE
5. The child gives a preparatory breath in anticipation of the resting tone and sings resting
tone on the correct pitch.
4. The child sings resting tone on the correct pitch.
3. The child sings resting tone on somewhat correct pitch; slides into or adjusts to correct
pitch.
2. The child is unable to find the correct resting tone pitch but sings within a whole step.
1. The child is unable to find the correct resting tone pitch.
TONAL BEHAVIOR: TONAL PATTERN IMITATION
5. The child sings all patterns accurately.
4. The child sings two patterns accurately.
3. The child sings one pattern accurately.
2. The child sings all patterns with correct melodic contour but with inaccurate pitches.
1. The child sings all patterns with incorrect melodic contour and inaccurate pitches.

	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
2
This is the final version of ECMBM, reflecting all revisions, that was used for the main study.
	
  

39	
  

Figure 1 (cont’d)

RHYTHM BEHAVIOR: CHANT PERFORMANCE
5. The child performs the entire chant with consistent tempo and with correct rhythmic
patterns.
4. The child performs the chant with consistent tempo and more than half but not all of the
chant is performed with correct rhythmic patterns.
3. The child performs the chant with consistent tempo and some but less than half of the
chant is performed with correct rhythmic patterns.
2. The child performs the chant with consistent tempo but with incorrect rhythmic patterns.
1. The child performs the chant with inconsistent tempo.
Note. If the child does not perform with consistent tempo, the child should still receive
a score of 1 in spite of correct rhythmic patterns.
RHYTHM BEHAVIOR: RHYTHM PATTERN IMITATION
5. The child chants all patterns in tempo and with accurate rhythms.
4. The child chants two patterns in tempo and with accurate rhythms.
3. The child chants one pattern in tempo and with accurate rhythms.
2. The child chants all patterns with inconsistent tempo but with somewhat accurate
rhythms.
1. The child chants all patterns with inconsistent tempo and with inaccurate rhythms.
RHYTHM BEHAVIOR: RHYTHM IMPROVISATION
5. The child performs all improvisatory patterns that correspond with the beat and uses a
rich variety of rhythmic vocabulary (more than two different patterns).
4. The child performs all improvisatory patterns that correspond with the beat but uses a
limited rhythmic vocabulary (one or two different patterns).
3. The child performs improvisatory patterns that correspond with the beat and may
imitate one or two teacher patterns.
2. The child performs improvisatory patterns without discernable rhythmic organization
and may imitate one or two teacher patterns.
1. The child does not perform improvisatory patterns but imitates teacher’s patterns.

	
  

40	
  

Figure 1 (cont’d)

RHYTHM BEHAVIOR: KEEPING BEAT
5. The child maintains a steady beat that corresponds with the beat of the song for all of
the song.
4. The child maintains a steady beat that corresponds with the beat of the song for most
but not all of the song.
3. The child maintains a steady beat that corresponds with the beat of the song for less
than half but not all of the song.
2. The child maintains a steady beat that does not correspond with the beat of the song.
1. The child does not perform a steady beat.
Note. Steady beat refers to either macrobeat or microbeat. However, combination of both
will not be considered as “performing a steady beat” for this dimension.
MOVEMENT BEHAVIOR
5. The child moves body in relation to music and uses flowing movements in locomotion
while also pulsing beat.
4. The child moves body in relation to music and uses flowing movements in locomotion.
3. The child moves body in relation to music and uses flowing movements in place.
2. The child moves body, and the movement is related to music.
1. The child moves body, but the movement is not related to music.

Construction of Assessment Activities
Music teachers plan a variety of activities in order to meet the goals of lesson objectives,
and the assessment should be based on what is actually taught in class. Many of the assessment
activities of ECMBM are based on a MLT approach (Gordon, 2013), which underpins the
curriculum that was typically taught in this school’s music classes. In order to design assessment
activities systematically, I determined the following guidelines.
First, each assessment activity should be an activity that is taught as a part of regular
music instruction. Also, it should enable the teacher to teach, observe, and assess simultaneously.

	
  

41	
  

Second, each assessment activity should use a minimum of verbal directions so that
children could generate musical behaviors naturally in response to teacher-initiated musical
behaviors, teacher-guided music activities, teacher-elicitation, and musical stimulus.
Third, each assessment activity should be simple enough for the children to understand
and so that they can successfully participate in it. Therefore, activities should incorporate
familiar directions, routines, or vocabulary that are used in class on a regular basis.
Description of Assessment Activities
The following assessment activities are in their original form and were constructed and
used first for the pilot study. I revised the activities that needed changes based on data collection
process during the pilot study and feedback provided from the validation panel members.
Detailed reasons for changes are further described in Chapter IV in Audio and Videotaping
Procedures and Piloting Assessment Activities.
Tonal Behavior: Singing Accuracy
Children sang one criterion song in minor tonality that had been taught to them in prior
classes (Appendix H). Therefore, the song was familiar enough for the children to sing it
successfully. The song was in the children’s singing range (middle C to the B above middle C)
and was a developmentally appropriate length.
There was some discrepancy among research findings regarding the use of text or neutral
syllables in singing. Several studies revealed that using songs with neutral syllables is
recommended when measuring singing accuracy (Goetze, 1985; Levinowitz, 1989), but Smale
(1987) suggested that having text does not necessarily interfere in measuring child’s singing
accuracy. Moreover, another study showed that children sing better with text than without
(Jacobi-Karna, 1996). The criterion song in this study contained simple and repetitious text in

	
  

42	
  

order to minimize the text interference while singing, because it seemed to help children
remember the length of the song and feel more comfortable when singing the song alone. In
addition, the notion of “singing” is, to an extent, culturally based, and typically singing has text.
During data collection, the children and I performed the song together as a review and to
establish the tonality and the meter. Then, I invited each child to participate in a singing game. I
walked around the room moving a leaf in a flowing motion while singing the first four measures
of the song. I then passed the leaf to a child to sing the subsequent four measures.
Tonal Behavior: Resting Tone
Children were familiar with singing the resting tone after I performed a dominant pitch
by the time that assessment took place, because this type of dominant-tonic activity was a routine
part of every music class. Children learned that the teacher’s dominant pitch invited them to sing
the tonic pitch in response. However, in order to reinforce and collect individual data, I gave the
following specific directions at the beginning of the activity: “When the butterfly lands on your
finger, please sing ‘bum.’” This verbal direction was sung on pitches within the established
tonality, or on the dominant pitch and “bum” was sung on the tonic pitch. The butterfly finger
puppet was used to elicit a resting tone response. I sang the song “Butterfly” (see Figure 2),
which was in natural minor tonality, and approached each child. The butterfly finger puppet was
used to elicit a resting tone response. The song was performed using a neutral syllable. At the
end of every four measures, where the asterisk is marked, I pause briefly and sang the dominant
pitch and landed the butterfly on the child’s arm or finger as a cue for the child to sing the resting
tone. However, due to a revision of the rating scale, it was necessary to capture child’s
“preparatory breath.” Therefore, for the main study, I placed a hula-hoop in the middle of the

	
  

43	
  

classroom and stationed the video camera up close. I asked each child to stand in the hula-hoop
and landed the butterfly to give a cue to perform the resting tone after my dominant pitch.
Figure 2
Criterion Song for Resting Tone Dimension

Butterfly

Tonal/Rhythm Behavior: Tonal/Rhythm Pattern Imitation
Children were rated on their singing of tonal patterns in both major and minor tonalities
and their chanting of rhythm patterns in both duple and triple meters. I used tonal and rhythm
patterns that Taggart (1994) selected according to difficulty level from Gordon’s (1976) Tonal
and Rhythm Patterns: An Objective Analysis (see Figure 3). Because pattern instruction was part
of music activities, the children were familiar with this routine. In MLT instruction, the teacher
usually introduces tonal or rhythm patterns to children, and, in most cases, children listen,

	
  

44	
  

respond to, or imitate teacher-initiated patterns. Usually, a major song is performed to provide a
context prior to major tonal pattern, and a minor song is performed to provide a context prior to
minor tonal pattern. For rhythm patterns, a duple meter chant or song is used to provide context
for duple meter pattern instruction, and a triple meter chant or song provides the context for
triple meter rhythm pattern instruction. For Rhythm Pattern Imitation, I chanted, “Can you echo
these patterns for me?” in duple or triple meter to establish the context and performed the
patterns for a child to echo.
Figure 3
Tonal and Rhythm Patterns Used for Tonal and Rhythm Pattern Imitation Dimensions

Note. Source from Taggart’s (1994) Validity study of “Audie”: A test of music aptitude for 3and 4-year-old children

	
  

45	
  

Rhythm Behavior: Chant Performance
Before the assessment, I asked the children to move their upper bodies, shoulders, arms,
and hands with continuous flow, placing a beanbag on various parts of their bodies, and
reviewed the chant, “Pancake” as a group (see Figure 4). Then, for the assessment, I asked six
children to sit in a circle and passed a beanbag on the macrobeats while I chanted, “Pass the
Beanbag” (see Figure 5) in a triple meter. Then, at the end of chant (i.e., last two measures of
the preparing chant), I chanted, “Can you do the Pancake Chant?” within a triple meter context to
the child who had the beanbag in his or her hands. In order to help the child to start the chant, I
chanted the upbeat and the first down beat with the child.
Figure 4
Criterion Chant for Chant Performance Dimension

Pancake

Figure 5
Preparing Chant for Chant Performance Dimension

Pass the Beanbag

	
  

46	
  

Rhythm Behavior: Rhythm Improvisation
To prepare the children for assessment activity, I chanted a four-beat rhythm pattern, and
the children improvised rhythmically for four subsequent beats. If a child improvised longer
than four beats, I allowed the child to finish his or her improvisation. Also, the tempo was
adjusted if the child chose to improvise in a different tempo. In a subsequent class period, I
reviewed by modeling a call-and-response for children. I showed two fists, and chanted two
patterns that were the same, and said, “same.” Then, I showed one fist, and one open palm,
modeled two patterns that were different and said, “not same.” This process was to help children
to differentiate pattern improvisation from pattern imitation. The children and I practiced “same”
and “not same” rhythm patterns as a group. Finally, I engaged individual children in the
assessment activity. For the assessment, I chanted “My Mother and Your Mother” from Music
Play (Valerio et al., 1998) and traded improvisations with each child. At the end of the chant, I
used this verbal cue, “Give me a different pattern, here we go,” and performed all four teacherinitiated patterns, one at a time, to which the child responded through improvisation for each
measure (see Figure 6).
Figure 6
Teacher-Initiated Patterns for Rhythm Improvisation Dimension

Rhythm Behavior: Keeping Beat
Children’s ability to keep beat was assessed using the following assessment modes: (1)
Gross Motor - tapping steady beat, (2) Manipulatives - playing steady beat on a drum, and (3)
Locomotor - walking to steady beat. The song “Love Somebody” (Appendix H) was used for

	
  

47	
  

assessment of keeping beat using gross motor movement. First, the children and I sang and kept
a steady beat by tapping our laps in bilateral motion to the beat. For data collection, I set up two
chairs back to back in the corner of the classroom, and two video cameras were angled to capture
the tapping motion up close. I asked if anyone wanted to keep beat for class. While the class
sang the song again, the two children sitting in the chairs kept beat by tapping on their laps. The
rest of the class kept beat quietly, patting their laps with spider shaped hands to make their
movement silent so that the sound could not be imitated by the children who were being
evaluated. During the pilot study, I noticed that children who were being evaluated could still
hear each other’s tapping, and a few children seemed to follow the tapping sound of the other
child. Moreover, I found that rating two children accurately at the same time was not possible.
Therefore, for the main study, children were assessed one at a time.
To measure beat keeping on an instrument, a drum was placed on the table in the corner
of classroom where the video camera was set up. The drum was set up facing the camera so that
accurate visual data could be captured and to minimize group influence (visual) as other children
kept the beat on their laps as well. Each child played the drum with two hands in bilateral
motion to the composed song, “Drumming” (Appendix H). So that I sustained a consistent
tempo throughout the assessment, I used a small metronome with blinking lights to set my
singing tempo.
The assessment to investigate whether children could walk to steady beat incorporated
pretend play. I tiptoed, pretending not to wake up a giant for the first part (A section) of “The
Sled” from Music Play (Valerio et al., 1998). Then, I pretended to be the giant for the second
part of the song (B section). For the assessment, all children sat in a circle holding a co-op band
(a large rubber stretchable band covered in fabric that, when laid on the floor, created a circle

	
  

48	
  

with a diameter of about 5 feet) so that each child’s walking could be videotaped within that
boundary. Then, each child walked around while the whole group sang the song. For the main
study, I did not use the co-op band because there was a large squared carpet, and it allowed the
children to move within that space. Also, my verbal direction changed, as children were
confused and unable to do with two different tasks without my moving with them. Therefore, I
asked children to put on a pair of giant shoes and stomp around but did not ask to tiptoe.
Movement Behavior
We sat in a circle and held the co-op band to form a pretend swimming pool. I moved
around in the circle and modeled swimming, using combinations of the movement elements:
flow, weight, space, time, and pulsations. Then I told children that they were going to jump into
the swimming pool and swim. I also explained that the swimming pool could be filled with
water, popcorn, or thick mud. I asked two children to swim for each assessment activity, and the
rest of the class kept beat with the co-op band to my singing. While singing the song, I varied
the tempo (moderate, slow, fast), dynamic (moderately loud, soft, loud), expression of the song,
and verbally cued the movement (e.g., “Swim in the mud pool”) in order to elicit interactions of
the movement elements: flow, weight, space, and time (Laban, 1971) and to initiate the
pulsations (Valerio et al., 1998). Due to major changes in the rating scale used for the main
study, I asked only one child to move and the rest of the class to move with flow sitting down. I
did not give any specific verbal cue; I asked the child to move however he/she wanted while
standing up.

	
  

49	
  

Design and Procedures
After I obtained an approval letter from the subjects’ school and IRB approval from
University’s Human Research Protection Program, I asked the Pre-K teachers to send a letter to
students’ homes requesting parental permission for their children to participate in the study.
Parents who granted consent for their child’s participation were asked if their child was willing
to participate in the study. All parents gave consent for their child’s participation in both the
pilot and the main study, and all were included in the sample. Because children were more likely
to respond freely in the presence of adults who are familiar and in a nonthreatening, familiar
context (Greenberg, 1976), the children’s musical behavior was assessed during their regular
music periods with their music teacher, who also was the researcher. Activities from the 40
minutes of each class were video- and/or audio-recorded for 6 weeks (two times per week) for
subsequent rating.
Since many of the children were attending preschool for the first time and were not
familiar with the preschool environment at the beginning of the school year, data collection did
not begin until several months into the school year. This allowed me to build a rapport with the
children and also for the children to become familiar with music activities. In addition to leading
music activities in the music classes prior to the beginning of data collection, I taught each
criterion song and chant to the children for 8 classes. The assessment occurred in the following 6
weeks.
Data Collection
All data collection for the study took place during the subjects’ regular music classes. All
movement activities were videotaped, and all vocal performances were audio recorded to collect
accurate sound data. During the videotaping, a number tag (visible in size) was attached to the

	
  

50	
  

subjects’ fronts so that subjects could be identified easily. Class 1 (n = 17) had white number
tags, and Class 2 (n = 19) had orange number tags to allow for group identification.
Rating Process
After data collection, audio and videotapes were edited in order to extract the assessment
activities for the judges’ convenience. To determine the interjudge reliabilities of the rating
scales, three independent judges rated each child’s musical behaviors with the researcherdeveloped rating scales and the rating sheets with instructions (Appendix F). For the child’s
individual vocal performances, such as singing and patterning, a voice recorder was used so that
judges could rate based on optimal sound quality, as video recording does not capture clear
sound. This also allowed for more confidentiality. I said each child’s assigned number into the
voice recorder before each performance to identify the children. The three judges watched
and/or listened to video/audio tapes for the rating process. Judges were selected based on their
experiences teaching early childhood music to infant/toddler and preschool children and
according to their familiarity with MLT-based early childhood music activities. In addition, I
rated the children’s musical behavior during class on assessment days, as the intent of ECMBM
is to use it in the context of music instruction. In addition to determining interjudge reliability, I
calculated intrajudge reliability using the ratings of the children’s performances that I did during
class and ratings that I did after class while listening to or watching the recorded data.
Data Analysis
Pearson Product Moment correlations were used to determine both intrajudge and
interjudge reliabilities of the rating scales. A high reliability coefficient for a rating scale is at
least .90 for a composite score and at least .70 for each individual dimension (Gordon, 2002). In
discussing the magnitude of reliabilities of the individual dimension in the results, I will

	
  

51	
  

describe .80 and above as high, .60 to .79 as moderate, and below .60 as low. To determine the
extent to which the distributions approximated normal distributions, which is evidence that a
measurement tool is functioning well with a specific group of students, I calculated the means
and standard deviations of the dimensions and composites for the three judges’ individual and
combined ratings. I compared these to the theoretical means and standard deviations.

	
  

52	
  

CHAPTER IV
THE PILOT STUDY
I conducted a pilot study in November of 2012 to practice and refine the assessment
techniques and instruments that I would be using in the main study and to determine the
preliminary reliabilities of the scale. Twelve randomly selected children from one intact class (N
= 20) who were not going to participate in the main study participated in the pilot study. The
subjects in the pilot study learned each criterion song and chant and participated in regular music
activities for the remaining 40 minutes of their class period, one time per week, for 8 weeks
during the instructional period prior to data collection. Data collection began the following week
and continued for 4 weeks. The subjects’ other classmates (n = 8) participated in the same music
activities, but I did not collect data for them. The data collection procedures and assessment
procedures were the same as or similar to the planned main study.
Five independent judges, including me, who are experts in early childhood music,
participated in the rating process. I provided the judges with a CD and a DVD of the children’s
performances and rating sheets with instructions for how to use the rating scale (Appendix F). In
addition to rating the students’ performances, judges also provided comments on the audio/video
quality, the rating scale, and the appropriateness of the researcher’s musical prompts in
collecting data from the children. For data analysis, Pearson Product Moment correlations were
used to determine interjudge reliabilities of the rating scales, and the means and standard
deviations were calculated and compared with the theoretical means and standard deviations.

	
  

53	
  

Findings
Audio and Videotaping Procedures
Prior to data collection, I assigned the children numbers to protect their identity during
the rating process. Then, I prepared rating cards to facilitate my rating process while collecting
data. On each card was one dimension of the rating scale and two columns, one for the
children’s identification numbers and the other for their scores on that dimension (Appendix G).
In spite of the initial plan to rate while collecting data so that I could establish intrajudge
reliability, the data collection process revealed that it was too difficult to collect the data and rate
at the same time at first. It was not only difficult to hold a voice recorder, the rating card, and a
pencil, but conducting activities and recording scores simultaneously hindered my ability to
determine appropriate ratings. As a result, I stopped recording scores in order to focus on
collecting the recordings of the children’s performances for subsequent analysis. However, as I
got comfortable recording the performances while conducting activities, I began exploring
recording each child’s score vocally into the voice recorder, which was revealed to be useful. I
recorded the score of each child’s performance vocally during the transition; that is, I recorded
the score while moving to the next child, getting props, or before continuing with group activities,
so that neither the child who performed nor other children heard the score. This method was
used to establish my intrajudge reliability for the main study but not for the pilot study. Only my
ratings from the recorded data were used to establish interjudge reliabilities with the other four
judges for the pilot study. When collecting visual data, I recorded the score on the rating card.
The children did not seem to be uncomfortable with the audio/videotaping procedures during
data collection process. They responded willingly and did not appear to be under any stress.

	
  

54	
  

There was a trial and error period as I developed the audio recording procedures.
Technical manipulation of the voice recorder hindered my ability to conduct activities in a fluid
manner while collecting data, resulting in wait-time between individual assessments. At the
beginning of the audio recording process, I tried to record each child’s performance separately; I
pushed the start button and the stop button before and after each child’s performance. However,
starting and stopping the recorder hindered my abilities to lead the assessment activities without
stopping. I found as a result of trial and error that recording a complete assessment activity on
one take, from beginning to end without stopping, and editing later seemed to facilitate my
ability to teach/conduct the assessment activities. In order to identify each child and leave
enough space for editing, I briefly paused for a moment between children’s performances, and I
voice-recorded each child’s number.
Since the intent of ECMBM is to assess children’s musical behavior in the context of
music activities, I made no attempt to control for peer-influence. However, peers seemed to
distract the children for certain dimensions, especially when keeping beat in the Gross Motor
mode. At first, two chairs were positioned back to back so that the children could not see each
other’s tapping motion. While this minimized the visual copying of beat, children could still
hear each other’s tapping, and a few children seemed to follow the tapping sound of the other
child. Moreover, I found that rating two children accurately at the same time was not possible.
Considering these discoveries, I decided to adjust the procedure by positioning only one
“tapping chair” in the corner rather than two. I asked the children in the group to tap silently,
whereas I asked the child who was being assessed to tap with sound so that each child’s tapping
was clear both visually and aurally, which would facilitate the rating process.

	
  

55	
  

At first, I let the class sing while one child was keeping beat. However, because children
as a group tend to rush the tempo of the song, I was worried that it could result inaccurate ratings
of the child’s performance. Therefore, I asked the class to “sing silently” to accompany their
silent tapping, while I performed the song and the individual child being rated sat in the chair and
tapped.
Piloting Assessment Activities
While I designed the assessment activities carefully to facilitate data collection, several
concerns emerged as I collected the data. I, as the teacher, tried to improvise and adjust activities
in response to the children in order to keep the setting as similar as possible to the usual class
setting. However, I found that the assessment activities for the purposes of data collection
needed to be more structured and consistent in terms of giving cues and providing introductory
labels. Providing a clear cue of when to start, either verbally or with a gesture in a consistent
manner was important. However, because I was engaged in multiple tasks, I was inconsistent in
giving cues and providing introductory labels. Furthermore, I forgot to remind children to do
“silent activities” when I was collecting individual data, so, at first, a few children had to stop
singing and start again. For example, because children had already learned the criterion song
“Autumn Leaves,” they began to sing the song along with the child who was being rated. I had
to remind children to be “silent trees” so that the child who was being assessed – “the singing
tree” - could sing by him/herself.
While collecting data for Tonal Behaviors (i.e., Singing Accuracy, Resting Tone, Tonal
Pattern Imitation), I found that I needed to get an accurate starting pitch using a recorder or tone
bell before the performance of each child so that I was consistent with the starting pitch, which

	
  

56	
  

allowed the children to sing in the same vocal range. For the main study, I used a recorder, as I
easily could put it back in my pocket after obtaining the first pitch.
I found that, for the Resting Tone dimension, the wordings of criterion 5 (The child gives
a preparatory breath in anticipation of the resting tone and sings resting tone on pitch) required
visual data in addition to audio data to capture child’s “preparatory breath.” Consequently, I
placed a hula-hoop in the middle of the classroom and stationed the video camera up close. I
asked each child to stand in the hula-hoop and landed the butterfly on the child’s arm or finger to
give a cue to perform the resting tone after my dominant pitch.
The most difficult assessment activities were related to movement. For Keeping Beat in
Locomotor mode, I asked the children to hold the co-op band to set a boundary so that the
children did not walk out of the camera view. However, this artificial boundary was not ideal, as
it seemed to change the quality of the children’s movement. For the main study, because the
classroom had a big square carpet, the camera was set up to capture any movement within that
space. This was ideal, as children were used to moving around on the square carpet. The
children’s distance from the camera did not matter, because I inserted each child’s identification
number on the left side of the screen using iMovie on Mac laptop when editing, so the judges
easily could identify the child who was being assessed.
In terms of instruction, I found that asking a child to walk to the song confused some of
the children, making them reluctant to move, as they had not explored “just walking” to music
previously without my moving with them. I usually moved with the children and modeled for
them. Therefore, for the main study, children specifically were asked to put on their giant shoes
and stomp around, and this activity was explored several times as a group and individually prior
to data collection.

	
  

57	
  

During the pilot study, for the Movement Behavior activity, a few children seemed to be
bothered by being pulled out of the group and some felt embarrassed to move, even though two
children were asked to move at the same time. Therefore, the activity had to be redesigned for
the main study. For the main study, the group was asked to move with flow sitting down, while
the child who was being assessed could move however he/she wanted while standing up. In
addition, because the pilot study revealed that it was difficult to see pulsations while two children
moved around each other, as occasionally they turned their backs to the camera or one child
completely blocked the view of the other child, only one child moved at a time and each child
was asked to face the camera for the main study.
Interjudge Reliabilities
The judges’ ratings for all dimensions of ECMBM were combined and correlated. The
resulting reliability coefficients ranged from .82 to .96, which are high.
Table 1.1
Pearson Correlations among Judges’ Composite Scores of All Dimensions
Judge 2
Judge 1
Judge 2

Judge 3

Judge 4

Judge 5

.82

.92

.91

.96

.84

.87

.86

.87

.93

Judge 3
Judge 4

.90

	
  
Tonal Behaviors
The interjudge reliability coefficients of individual judge’s composite scores for Tonal
Behaviors – Singing Accuracy, Resting Tone, and Tonal Pattern Imitation (Major/Minor) – were
high, ranging from .83 to .96 (see Table 1.2). As might be expected, the reliability coefficients

	
  

58	
  

for combined dimensions of the rating scales were higher than for the individual dimensions of
the rating scale.
Table 1.2
Pearson Correlations among Judges’ Composite Scores of Tonal Behavior Dimensions
Judge 2
Judge 1

Judge 3

Judge 4

Judge 5

.89

.94

.90

.96

.88

.89

.93

.88

.94

Judge 2
Judge 3
Judge 4

.83

While Singing Accuracy (r = .82 - .96) and the Resting Tone dimensions (r = .83 - .93)
resulted in a range of high reliabilities (see Tables 1.3 and 1.4), the individual dimensions of
Major Pattern Imitation (r = .72 - .86), and Minor Pattern Imitation (r = .74 - .94) resulted in
ranges of moderate to high reliabilities (see Tables 1.5 and 1.6).
Table 1.3
Pearson Correlations among Judges’ Ratings on Singing Accuracy Dimension
Judge 2
Judge 1
Judge 2

Judge 3

Judge 4

Judge 5

.93

.92

.88

.85

.87

.95

.87

.82

.96

Judge 3
Judge 4

	
  

.82

59	
  

Table 1.4
Pearson Correlations among Judges’ Ratings on Resting Tone Dimension
Judge 2
Judge 1

Judge 3

Judge 4

Judge 5

.90

.93

.89

.92

.83

.89

.93

.88

.88

Judge 2
Judge 3
Judge 4

.92

The reliability coefficients for Minor Pattern Imitation (r = .74 - .94) represented a wider
range and somewhat higher reliabilities than was true for the Major Pattern Imitation dimension
(r = .72 - .86). 	
  
Table 1.5
Pearson Correlations among Judges’ Ratings on Major Pattern Imitation Dimension
Judge 2
Judge 1

Judge 3

Judge 4

Judge 5

.73

.86

.78

.85

.84

.83

.82

.74

.85

Judge 2
Judge 3
Judge 4

.72

	
  
Table 1.6
Pearson Correlations among Judges’ Ratings on Minor Pattern Imitation Dimension
Judge 2
Judge 1
Judge 2

Judge 3

Judge 4

Judge 5

.74

.81

.90

.76

.78

.74

.81

.94

.94

Judge 3
Judge 4

.84

	
  

	
  

60	
  

When the ratings of both Major and Minor Pattern Imitation dimensions were combined,
the reliability coefficients ranged from .76 to .93, with the lowest interjudge reliability with .76
being an outlier. The next lowest interjudge reliability was .84.	
  
Table 1.7
Pearson Correlations among Judges’ Composite Scores of Tonal Pattern Imitation Dimensions
Judge 2
Judge 1

Judge 3

Judge 4

Judge 5

.76

.87

.85

.89

.89

.85

.93

.91

.91

Judge 2
Judge 3
Judge 4

.84
Rhythm Behaviors

Individual judge’s composite scores for Rhythm Behavior dimensions – Chant
Performance, Rhythm Pattern Imitation (Duple/Triple), Rhythm Improvisation, and Keeping
Beat (Gross Motor, Manipulatives, Locomotor) – also resulted in a range of high reliabilities (r
= .80 - .93) (see Table 1.8), although not as high as those for the Tonal Behaviors.
Table 1.8
Pearson Correlations among Judges’ Composite Scores of Rhythm Behavior Dimensions
Judge 2
Judge 1
Judge 2

Judge 3

Judge 4

Judge 5

.81

.81

.89

.93

.81

.80

.86

.92

.84

Judge 3
Judge 4

	
  

.93

61	
  

When examining the reliability coefficients of the individual dimensions, it is evident that
there were some disagreements among judges. There was a wide range of interjudge reliability
coefficients, ranging .41 - .93 for Chant Performance (see Table 1.9); a narrower range of .68 .91 for Duple Pattern Imitation (see Table 1.10); a wide range of .20 - 1.00 for Triple Pattern
Imitation (see Table 1.11); and a narrower range of .54 - .96 for Rhythm Improvisation (see
Table 1.13).
As seen in Table 1.9, the reliability coefficients for Chant Performance mostly were
moderate to high (r = .60 - .93), and a single low reliability coefficient was found between the
scores of Judges 1 and 2 (r =. 41). The Chant Performance dimension resulted in a wide range of
reliability coefficients, with most being moderate (n = 6), one being low, and several being high
(n = 3).
Table 1.9
Pearson Correlations among Judges’ Ratings on Chant Performance Dimension
Judge 2
Judge 1
Judge 2

Judge 3

Judge 4

Judge 5

.41

.60

.70

.68

.76

.70

.71

.80

.86

Judge 3
Judge 4

.93

	
  
As can be seen in Table 1.11, reliability coefficients for Triple Pattern Imitation were
extremely varied, ranging from .20 to 1.00. While there was a perfect interjudge reliability
coefficient between Judges 1 and 5, the scores of all other judges resulted in low (n = 5) to
moderate (n = 4) reliabilities. The reliability coefficients for the composite scores of the Rhythm
Pattern Imitation dimensions (see Table 1.12) resulted a range of moderate to high (r = .63 - .91)
reliabilities.

	
  

62	
  

Table 1.10
Pearson Correlations among Judges’ Ratings on Duple Pattern Imitation Dimension
Judge 2
Judge 1

Judge 3

Judge 4

Judge 5

.70

.90

.89

.79

.77

.68

.91

.88

.90

Judge 2
Judge 3
Judge 4

.72

Table 1.11
Pearson Correlations among Judges’ Ratings on Triple Pattern Imitation Dimension
Judge 2
Judge 1

Judge 3

Judge 4

Judge 5

.67

.43

.74

1.00

.20

.55

.67

.43

.43

Judge 2
Judge 3
Judge 4

.74

Table 1.12
Pearson Correlations among Judges’ Composite Scores of Rhythm Pattern Imitation Dimensions
Judge 2
Judge 1
Judge 2

Judge 3

Judge 4

Judge 5

.74

.70

.91

.86

.63

.76

.88

.76

.83

Judge 3
Judge 4

.91

The Rhythm Improvisation dimension resulted reliability coefficients ranging from .54
to .96. The low and moderate reliability coefficients were between Judge 3 and all other judges
(see Table 1.13). Except for Judge 3, there was fairly strong agreement among judges.

	
  

63	
  

Table 1.13
Pearson Correlations among Judges’ Ratings on Rhythm Improvisation Dimension
Judge 2
Judge 1

Judge 3

Judge 4

Judge 5

.84

.54

.80

.96

.56

.71

.77

.68

.62

Judge 2
Judge 3
Judge 4

.92

As can be seen in Table 1.14, the reliability coefficients of composite scores on the
Keeping Beat dimensions ranged moderate to high (r = .69 - .95). However, the individual
dimensions (i.e., Gross Motor and Locomotor) in this category resulted a wide range of
interjudge reliabilities. The lowest interjudge reliabilities for Keeping Beat were .35 for Gross
Motor, .49 for Locomotor, and .77 for Manipulatives. The highest interjudge reliabilities for
Keeping Beat performance were .87 for both Gross Motor and Locomotor, and .98 for
Manipulatives.
Table 1.14
Pearson Correlations among Judges’ Composite Scores of Keeping Beat Dimensions
Judge 2
Judge 1
Judge 2

Judge 3

Judge 4

Judge 5

.80

.80

.87

.87

.74

.71

.69

.84

.79

Judge 3
Judge 4

.95

The reliability coefficients for Keeping Beat: Gross Motor ranged from low to high (r
= .35 - .87). However, as can be seen in Table 1.15, there were more moderate reliability
coefficients (n = 6) than either low (n = 2) or high (n = 2) reliability coefficients. In comparison

	
  

64	
  

to the reliabilities of Keeping Beat: Gross Motor, interjudge reliability coefficients for using
manipulatives resulted in mostly high interjudge reliability coefficients except for the
coefficients between Judges 3 and 5 (r = .77), Judge 3 and 4 (r = .79), and Judge 1 and 5 (r
= .79) all of which were approaching high.
Table 1.15
Pearson Correlations among Judges’ Ratings on Keeping Beat: Gross Motor Dimension
Judge 2
Judge 1

Judge 3

Judge 4

Judge 5

.35

.59

.70

.64

.77

.76

.66

.87

.82

Judge 2
Judge 3
Judge 4

.79

Table 1.16
Pearson Correlations among Judges’ Ratings on Keeping Beat: Manipulatives Dimension
Judge 2
Judge 1
Judge 2

Judge 3

Judge 4

Judge 5

.82

.85

.83

.79

.81

.93

.87

.79

.77

Judge 3
Judge 4

.98

The reliability coefficients for Keeping Beat: Locomotor ranged from low to high (r = .49
- .87). Similar to Keeping Beat: Gross Motor dimension, there were more moderate reliability
coefficients (n = 6), than either high (n = 3) and low (n = 1) reliability coefficients.

	
  

65	
  

Table 1.17
Pearson Correlations among Judges’ Ratings on Keeping Beat: Locomotor Dimension
Judge 2
Judge 1

Judge 3

Judge 4

Judge 5

.67

.79

.68

.83

.66

.49

.78

.69

.84

Judge 2
Judge 3
Judge 4

.87
Movement Behavior

The reliability coefficients for the Movement Behavior dimension mostly were low (r
= .20 - .54). However, moderate reliabilities were reported between Judges 3 and 4 and Judges 3
and 5 (r = .60), and a high reliability was reported between Judges 4 and 5 (r = .85).
Table 1.18
Pearson Correlations among Judges’ Ratings on Movement Behavior Dimension
Judge 2
Judge 1
Judge 2

Judge 3

Judge 4

Judge 5

.20

.27

.25

.43

.25

.54

.43

.60

.60

Judge 3
Judge 4

.85
Means and Standard Deviations

Composite scores of all dimensions resulted in much higher observed means than would
be expected in relation to the theoretical mean (TM = 36). However, the observed standard
deviations were much lower than the theoretical standard deviation (TSD = 8). In other words,
the rating scale in its entirety may have been a bit too easy for the children, and the scores were
closer together than would be true in a normal distribution.

	
  

66	
  

Table 2.1
Means and Standard Deviations of Composite Scores of All Dimensions by Judges
M

SD

Judge 1

42.08

4.94

Judge 2

42.08

6.56

Judge 3

43.75

5.80

Judge 4

40.67

4.60

Judge 5

41.17

5.73

Note. The score range of all combined dimensions (N = 12) is 12 – 60.
The observed means and standard deviations for the Tonal Behavior dimensions can be
seen in Table 2.2. For the Singing Accuracy dimension, the theoretical mean was 3 and the
theoretical standard deviation was .67. The observed means of Judges 2 and 3 were higher than
the theoretical means, whereas the means of the other judges approximated the theoretical means.
Except for the standard deviation for Judge 1, all standard deviations were greater than the
theoretical standard deviations. For the Resting Tone dimension, the means overall were lower
than the theoretical mean of 3, and the observed standard deviations all were higher than the
theoretical standard deviations. For the pattern imitation dimensions combined, the observed
means all were higher than the theoretical mean of 6, and the observed standard deviations all
were higher than the theoretical standard deviations. Overall, the scores were higher and more
broadly dispersed for these dimensions than would be true in a normal distribution.

	
  

67	
  

Table 2.2
Means and Standard Deviations of Tonal Behavior Dimensions

Judge 1

Singing Accuracy
(TM=3; TSD=.67)
M
SD
3.33
.65

Resting Tone
(TM=3; TSD=.67)
M
SD
2.25
1.22

Pattern Imitation
(TM=6; TSD=1.33)
M
SD
6.33
1.61

Judge 2

3.83

1.19

2.75

1.06

7.42

2.31

Judge 3

4.25

1.06

2.50

1.09

7.17

2.17

Judge 4

2.83

.94

2.42

.90

6.42

1.62

Judge 5

2.75

.97

2.58

1.00

6.75

1.96

Note. Score range: Singing Accuracy and Resting Tone (1-5); Pattern Imitation (2-10)
For the Chant Performance dimension (see Table 2.3), the means approximated the
theoretical means, as did the standard deviations. The distribution of this dimension was close to
normal.
Table 2.3
Means and Standard Deviations of Rhythm Behavior Dimensions

Judge 1

Chant Performance
(TM=3; TSD=.67)
M
SD
3.42
.67

Pattern Imitation
(TM=6; TSD=1.33)
M
SD
7.42
.79

Keeping Beat
(TM=9; TSD=2)
M
SD
12.75
2.01

Judge 2

2.83

.94

7.58

.79

11.58

1.68

Judge 3

3.33

.99

7.75

.87

13.08

1.78

Judge 4

3.33

.65

7.17

1.03

12.17

1.75

Judge 5

3.25

.75

7.42

.79

11.83

2.21

Note. Score range: Chant Performance (1-5); Pattern Imitation (2-10); Keeping Beat (3-15)

	
  

68	
  

I combined the Duple and Triple Pattern Imitation scores into a single composite
dimension (see Table 2.3). These observed composite means were higher than the theoretical
means but the observed standard deviations were mostly lower than expected in comparison to
the theoretical standard deviations. In other words, the scores were somewhat clumped toward
the top of the distribution. This was the case because the Duple Pattern Imitation resulted very
high observed means, which was not the case for Triple Pattern Imitation (see Table 2.4).
However, the observed standard deviations for Duple Pattern Imitation approximated the
theoretical standard deviations, whereas the observed standard deviations for Triple Pattern
Imitation were considerably smaller than expected.
Table 2.4
Means and Standard Deviations of Duple and Triple Pattern Imitation Dimensions
Duple Pattern Imitation
(TM=3; TSD=.67)
M
SD

Triple Pattern Imitation
(TM=3; TSD=.67)
M
SD

Judge 1

4.33

.65

3.08

.29

Judge 2

4.42

.67

3.17

.39

Judge 3

4.42

.52

3.33

.49

Judge 4

4.17

.84

3.00

.43

Judge 5

4.33

.65

3.08

.29

Note. The score range for a single dimension is 1-5.
The observed means for composite scores of the Keeping Beat dimensions were
substantially higher than the theoretical means and the observed standard deviations were close
to the theoretical standard deviations (see Table 2.3). In other words, the distribution was
skewed to the left. As expected in light of the composite means and standard deviations, when

	
  

69	
  

examining individual dimensions, the observed means consistently were higher than the
theoretical means (see Table 2.5), and the observed standard deviations were somewhat high.
Table 2.5
Means and Standard Deviations of Keeping Beat Dimensions by Mode

Judge 1

Gross Motor
(TM= 3; TSD = .67)
M
SD
4.58
.67

Manipulatives
(TM= 3; TSD = .67)
M
SD
4.17
.94

Locomotor
(TM= 3; TSD = .67)
M
SD
4.00
1.04

Judge 2

4.33

.65

3.58

1.17

3.67

.78

Judge 3

4.58

.67

4.58

.67

3.92

1.00

Judge 4

4.42

.79

3.92

1.31

3.83

1.03

Judge 5

4.33

.78

3.83

1.27

3.67

1.16

Note. The score range for each dimension is 1-5.
For the Rhythm Improvisation dimension, the means were slightly higher than the
theoretical means, and the observed standard deviations were higher than the theoretical standard
deviations. In other words, the distribution was skewed slightly to the left and the distribution
was platykurtic.
Table 2.6
Means and Standard Deviations of Rhythm Improvisation Dimension
M

SD

Judge 1

3.83

1.03

Judge 2

3.25

1.22

Judge 3

3.00

1.48

Judge 4

3.92

1.08

Judge 5

3.92

1.00

Note. The score range is 1-5. TM = 3; TSD = .67.

	
  

70	
  

For the Movement Behavior dimension, the observed means and standard deviations
were highly variable. The scores of Judges 2 and 3 seem to be problematic for the Movement
Behavior dimension, as they resulted in substantially higher observed means and standard
deviations than those of the other judges. The judges did not appear to be using the dimension in
a consistent way.
Table 2.7
Means and Standard Deviations of Movement Behavior Dimension
M

SD

Judge 1

3.33

.65

Judge 2

3.83

1.19

Judge 3

4.25

1.06

Judge 4

2.83

.94

Judge 5

2.75

.97

Note. The score range is 1-5; TM = 3; TSD = .67
Discussion of the Pilot Study
The results indicated that the preliminary version of ECMBM is useful for measuring
children’s tonal behavior, as high interjudge reliabilities were preponderant. However, it could
be invalid for measuring children’s rhythm behavior, as reliability coefficients ranged from low
to high, and movement behavior, as reliability coefficients mostly were low. In addition to the
reported reliability coefficients, judges’ comments and questions provided a window to
understand better the results of Rhythm and Movement Behavior dimensions.
The disagreements between judges’ ratings on the Rhythm and Movement Behavior
dimensions may be due to several reasons. First, the wordings of criteria were confusing.
Second, some children’s musical behavior did not fit into any of the descriptions in the criteria

	
  

71	
  

due to data collection procedures. Third, judges’ tonal judgments could have been more accurate
than their rhythm judgments, as the reliability coefficients of Rhythm Behaviors were low and
dispersed when compared to those of the Tonal Behaviors.
The judges provided feedback that a few descriptors within the criteria were not clear.
For example, they identified word choices such as “randomly” and “purposefully” in Chant
Performance dimension; “anticipate” in Resting Tone dimension; and “purposefully” in Rhythm
Improvisation dimension as unclear. In addition, comments revealed that some judges counted a
combination of macrobeat changing to microbeat or vice versa as keeping steady beat, whereas
others did not. Without a clear indication as to whether a combination of macrobeat and
microbeat should be considered to be steady beat, there may have been disagreements, resulting
in a wide range of reliabilities. Therefore, because the reliability of a rating scale depends upon
the clarity of its descriptive criteria, I revised the scale to provide definitions of unfamiliar words
and additional information (Gordon, 2002) to clarify the misunderstandings that were identified
by the judges.
The audio/video data revealed that I was inconsistent in giving instructions. For the
Movement dimension, I sometimes asked children to pulse, sometimes did not, and sometimes
gave different instruction, such as “splash your fingers” or “splash around.” Therefore, some
children might not have known that they were supposed to pulse, and some children could have
exhibited behavior that did not resemble one of the criteria. In addition to the inconsistency in
my assessment instructions, my musical prompts and/or introductory labels were inconsistent as
well. For a few of the children’s resting tone performances, only my dominant pitch and the
child’s resting tone pitch were recorded without providing previous singing context for the
judges. A few judges commented that it was difficult to rate a child’s resting tone performance

	
  

72	
  

in the absence of previous tonal context, which was the song before the teacher’s dominant pitch.
Therefore, for the main study, I made sure that all individual data included the song and the
teacher’s dominant pitch following the song, as well as the child’s resting tone response.
The pilot study also revealed that the children’s musical behaviors themselves as well as
the materials that I provided to the judges could be the causes of the low reliabilities. The
children were not prepared to exhibit certain musical behaviors. Because children only were
familiar with moving as a group, moving alone or in pairs could have made them feel self
conscious about their movement. ECMBM is constructed for a use within the context of music
activities, so data was collected in a way that reflected a regular music class setting as much as
possible. Consequently, when a few of the children were hesitant to move alone or in pairs, I
repeated the activity so that the children could feel more comfortable. However, because I
included the repetitions in the materials that I provided to the judges, they were confused. The
judges were not sure whether to use one performance or both performances in their ratings.
Therefore, in the main study, I had one child moving alone standing up, while the other children
moved sitting down to make the child who was moving alone feel comfortable, and I only
included the second take of two trials.
In addition to the aforementioned reasons for the low and moderate reliabilities, the small
number of children (N = 12) participating in the pilot study may have contributed to the low to
moderate reliabilities as well. For the actual study, 36 children participated, and the results
include the ratings of three independent judges instead of five, as was the case in the pilot study.
I used five in the pilot study so that I could gather rich feedback about the scale so that it could
be improved. This was less important for the final study, which had the focus of seeing how well
the scales worked rather than gathering information for improvement of the scales.

	
  

73	
  

In summary, as a result of the pilot study, I revised the rating scale in terms of wording
and redesigned the assessment procedures so that my musical prompts and requests were
consistent when eliciting performances from the children. Moreover, in addition to learning the
criterion song and chant as preparation for gathering data, the children who participated in the
main study also explored moving individually so that they were not uncomfortable moving alone
during the data collection process.

	
  

74	
  

CHAPTER V
FURTHER REVISIONS OF THE RATING SCALE
Content Validation of the ECMBM
The validity of a measurement indicates the extent to which the test measures what it is
supposed to measure. In the absence of external criteria for use in establishing concurrent
validity, I focused on establishing content validity based upon the feedback of the expert judges
who participated in the rating process for the pilot study as well as through consultation with
university professors whose expertise is in early childhood music and/or music education
measurement. The process of establishing the content validity of ECMBM is discussed in this
chapter. Because the assessment activities and the rating scales were constructed to conform to
MLT-based learning, seven early childhood music and measurement experts who are familiar
with MLT were asked to provide feedback for the content validity of ECMBM.
The validation process comprised two parts. First, the validation panel members were
asked to arrange the randomly ordered criteria into a sequential order for each dimension. If they
were able to do so in a consistent way, it would provide evidence that the scale reflected an
appropriate developmental sequence and that the criteria were appropriately descriptive of
children’s musical behaviors. They also provided their feedback to me in the form of written
comments. The panel of experts also looked at the assessment activities to determine whether
they were appropriate for use in an early childhood music classroom context. No statistical
treatment was applied, as content validity ultimately is subjective in nature. However,
descriptive results concerning the agreement to the researcher-established sequence of criteria	
  
and validation panel’s comments and concerns are reported below.

	
  

75	
  

Sequencing of the Criteria
For Pattern Imitation, both Tonal and Rhythm, and the Keeping Beat dimensions, 100
percent of the validation panel members agreed with the sequence of the criteria that I had
established. For the dimensions of Singing Accuracy, Resting Tone, and Chant Performance,
panel members’ agreement with the sequence of criteria was 94 percent (see Tables 3.1 - 3.3). In
comparison to the 100 percent and 94 percent agreement in these dimensions, the agreement rate
(54 percent) of judges for Rhythm Improvisation was low (see Table 3.4). For the Movement
dimension, although the agreement rate of validation panel members who responded was 100
percent, one validation panel member believed that the description of the criteria did not measure
what it was supposed to measure and that it was invalid in terms of content. Therefore, that
panel member did not provide responses.
Table 3.1
Correct Responses for Singing Accuracy Dimension
Criteria
5
4
3
2
1

RSC
D
C
A
E
B

PM1
D
C
A
E
B

PM2
D
C
E
A
B

PM3
D
C
A
E
B

PM4
D
C
A
E
B

PM5
D
C
A
E
B

PM6
D
C
A
E
B

PM7
D
C
A
E
B

Note. (1) RSC – The researcher-established sequence of criteria (2) PM – Panel Member
Table 3.2
Correct Responses for Resting Tone Dimension
Criteria
5
4
3
2
1

	
  

RSC
E
A
C
D
B

PM1
E
A
C
D
B

PM2
A
E
C
D
B

PM3
E
A
C
D
B

PM4
E
A
C
D
B

76	
  

PM5
E
A
C
D
B

PM6
E
A
C
D
B

PM7
E
A
C
D
B

Table 3.3
Correct Responses for Chant Performance Dimension
Criteria
5
4
3
2
1

RSC
C
B
D
A
E

PM1
C
B
D
A
E

PM2
C
B
A
D
E

PM3
C
B
D
A
E

PM4
C
B
D
A
E

PM5
C
B
D
A
E

PM6
C
B
D
A
E

PM7
C
B
D
A
E

PM5
C
D
E
A
B

PM6
C
E
D
B
A

PM7
C
E
D
B
A

Table 3.4
Correct Responses for Rhythm Improvisation Dimension
Criteria
5
4
3
2
1

RSC
C
D
E
A
B

PM1
C
E
D
A
B

PM2
C
E
D
A
B

PM3
C
E
D
A
B

PM4
C
E
D
A
B

For the Rhythm Improvisation dimension, Panel Member 6 and 7 believed that “imitating
the pattern” was a higher-level of musical behavior developmentally than “performing
improvisatory patterns without corresponding with the beat.” Moreover, as can be seen in Table
3.4, all panel members except Panel Member 5 perhaps believed that the word “purposefully”
(see Figure 7) represented positive and higher achievement behavior and “limited” inferred
negative and lesser achievement. This may have been due to an unclear criterion description.
Figure 7
Criteria 3 and 4 of the Original Rhythm Improvisation Dimension
4
3

	
  

The child performs improvisatory patterns within the meter but uses a limited rhythmic
vocabulary and may imitate a few patterns as a part of the improvisation.
The child purposefully improvises patterns that corresponds with beat and may imitate
a few patterns as part of the improvisation.

77	
  

Panel Members’ and Judges’ Comments
The second part of the content validation process was based on comments and feedback
provided from members of the validation panel. In addition, the pilot study and judges’
comments and questions were considered in revising the scale. In this section, I will provide a
detailed analysis of comments, feedback, and questions from the validation panel and judges that
I used when revising the preliminary ECMBM.
Tonal Behavior: Singing Accuracy
The original rating scale for Singing Accuracy and its revision can be seen in Figure 8.
The main concern expressed about this dimension was the need to clarify what the scale intends
to assess. The intent of this scale was to measure singing accuracy in terms of melodic contour
and pitch. The original rating scale included wording about the use of singing voice for those
children who mostly chant because they have not found their singing voice. However, by
including child’s use of singing voice, the scale was measuring two different constructs, singing
voice and pitch accuracy, making the rating scale less valid in terms of its content. Therefore, I
excluded the description, “does not use singing voice” in criterion 1.
In addition to the concern about measuring two constructs in one dimension, panel
members questioned the validity of measuring both melodic contour and pitch. The Singing
Accuracy dimension requires judges to consider both melodic contour and pitch. While the
judges might face difficulty in attending to both melodic contour and pitch simultaneously, I
believe that correct pitch has to be measured in conjunction with correct melodic contour. That
is, listening for correct pitches also requires one to listen for correct melodic contour. In order to
emphasize what this rating scale measures and to avoid possible confusion, I changed the
wording from “pitch level,” to “correct pitch.”

	
  

78	
  

The judges from the pilot study expressed that there was subjectivity in deciding length
of accurate singing. Because children sang four measures of a total of eight measures, the
wording “parts of the song,” or “most of the song” required subjective decisions depending on
the judges’ discretions. That is, judges’ ratings could be subjective as “parts” could be
represented by measures, tonal patterns, or individual pitches. Although Singing Accuracy
resulted in high interjudge reliabilities (r = .82 – .96), in order to improve the rating process and
increase the interjudge reliabilities further, I revised the criteria to focus on rating correct pitches
rather than length. Therefore, criterion 1 describes singing with no contour and pitch, and
subsequent higher criteria describe singing with correct contour but pitch performance differs in
terms of its degree.
Figure 8
The Original and Revised Rating Scales for Singing Accuracy Dimension
TONAL BEHAVIOR: SINGING ACCURACY
ORIGINAL
5. The child sings the entire song on the correct pitch level and melodic contour.
4. The child sings most of the song on the correct pitch level and melodic contour.
3. The child sings parts of the song on the correct pitch level and melodic contour.
2. The child sings with some correct melodic contour but on the wrong pitch level.
1. The child does not use singing voice or sings in a monotone; tonal context and
melodic contour are absent.
REVISED
5. The child sings the entire song with correct melodic contour and on the correct
pitches with good intonation.
4.	
  The child sings the song with correct melodic contour but with a few incorrect pitches
or poor intonation.
3. The child sings the song with correct melodic contour but on the wrong pitches.
2. The child sings the song with some correct melodic contour but on the wrong pitches.
1. The child sings with a lack of tonal context and melodic contour is absent.

	
  

79	
  

Tonal Behavior: Resting Tone
The original rating scale for Resting Tone and its revision can be seen in Figure 9. Four
members of the validation panel and all of the pilot study judges suggested that the word
“anticipate” in criterion 5 be defined clearly, as it did not seem to describe an observable
behavior. Although the word “anticipate” is an understood term for early childhood music
teachers who are familiar with MLT-based learning (Valerio et al., 1998), and teachers learn to
intuitively observe when a child is anticipating a resting tone, the validity of this part of the scale
was questioned due to its abstract description. Based on suggested rewording from two
validation panel members, I revised criterion 5 to include “preparatory breath” as an observable
measure of a child’s anticipation of the resting tone.
The measurement of degree also emerged as a concern for this dimension. The judges
and one validation panel member suggested that the word “somewhat” be revised to be more
specific. In order to facilitate judges’ arrival at objective decisions, a pitch boundary was
established. Based on one panel member’s suggestion, I added the more specific description,
“slides into or adjusts to the correct pitch” to replace “somewhat correct pitch.”
As was true for the Singing Accuracy dimension, I reconsidered the use of the description
“use of singing voice,” as the dimension was designed to measure whether a child can sing the
correct pitch for the resting tone, rather than how well the child is able to use singing voice. In
the revision, criteria 1 and 2 both describe the behavior of a child not being able to find the
correct resting tone pitch, but they differ in terms of distance from the correct pitch; for criterion
3, a child adjusts his or her pitch to the resting tone, even though it is not correct in the beginning.
Criteria 4 and 5 both describe the behavior of a child who sings the correct pitch, but criterion 5
gathers evidence of a child’s ability to audiate (Gordon, 2013) – thinking ahead of the resting

	
  

80	
  

tone in relation to teacher’s dominant pitch - rather than simply producing correct resting tone
randomly in response to teacher’s dominant pitch. While the child may perform the correct
resting tone without audiating, a child who audiates is more likely to perform the accurate pitch.
Figure 9
The Original and Revised Rating Scales for Resting Tone Dimension
TONAL BEHAVIOR: RESTNG TONE
ORIGINAL
5. The child anticipates resting tone, sings resting tone on the correct pitch.
4. The child sings resting tone on the correct pitch.
3. The child sings resting tone on somewhat correct pitch.
2. The child uses singing voice but is unable to find the correct pitch.
1. The child does not use singing voice.
REVISED
5. The child gives a preparatory breath in anticipation of the resting tone and sings
resting tone on the correct pitch.
4. The child sings resting tone on the correct pitch.
3. The child sings resting tone on somewhat correct pitch; slides into or adjusts to
correct pitch.
2. The child is unable to find the correct resting tone pitch but sings within a whole step.
1. The child is unable to find the correct resting tone pitch.
Tonal Behavior: Tonal Pattern Imitation
The original rating scale for Tonal Pattern Imitation and its revision can be seen in Figure
10. As with all other tonal dimensions, one member of the validation panel voiced a concern that
the dimension was measuring two different constructs – contour/pitch and singing voice.
Another validation panel member questioned the validity of counting the number of patterns
when measuring pattern imitation, instead of accounting for breathing and moving, which is
important to tonal/rhythm development according to MLT-based learning. Although the stages
of preparatory audiation (Valerio et al., 1998) address this issue, and breathing and moving are
important to observe in children’s responses to teacher-initiated patterns, I decided not to revise
	
  

81	
  

the scale based on these suggestions for several reasons. First, in regard to counting the number
of correct patterns, I used patterns that were selected based on audiation rather than performance
difficulty (Taggart, 1994); therefore, quite a few children who could not sing the easy pattern
were able to sing the medium pattern, and those who could not sing the medium pattern could
sing the difficult pattern, because sometimes a pattern is difficult to sing, even though it is easy
to audiate. Second, because this dimension measures children’s ability to accurately imitate
patterns that are performed by the teacher, its main focus is on children’s performance of
accurate patterns and not breathing and movement. I was concerned that, if I added breathing
and movement to the rating scale in conjunction with pattern accuracy, it would compromise the
judges’ ability to listen for melodic contour and pitch, as it would require the judges to assess
multiple variables.
Figure 10
The Original and Revised Rating Scales for Tonal Pattern Imitation Dimension
TONAL BEHAVIOR: TONAL PATTERN IMITATION
ORIGINAL
5. The child sings three patterns accurately.
4. The child sings two patterns accurately.
3. The child sings one pattern accurately.
2. The child sings the patterns with correct melodic contour but with inaccurate pitches.
1. The child sings the patterns with incorrect contour and incorrect pitches; sings in
monotone.
REVISED
5. The child sings all patterns accurately.
4. The child sings two patterns accurately.
3. The child sings one pattern accurately.
2. The child sings all patterns with correct melodic contour but with inaccurate pitches.
1. The child sings all patterns with incorrect melodic contour and inaccurate pitches.

	
  

82	
  

Rhythm Behavior: Chant Performance
The original rating scale for Chant Performance and its revision can be seen in Figure 11.
As with the Resting Tone dimension, one validation panel member and two pilot study judges
suggested revising the description of non-observable behavior (i.e., purposefully and randomly).
In addition, members of the validation panel advised that words that described length (i.e., parts
and most) be revised for more specificity. The validation panel’s comments aligned with judges’
comments from the pilot study, as they also had difficulty deciding between “parts” and “most.”
I agreed that specific indication of length could increase reliability. Therefore, I established a
specific boundary for criterion 3 as “less than half of the chant” and as “more than half of the
chant” for criterion 4.
One member from the validation panel and one judge from the pilot study voiced a
concern that this dimension was conflating two constructs: tempo and rhythm in one rating scale.
I decided that tempo should remain in the criteria, because correct rhythm is not really correct
unless it is performed in the context of a consistent tempo. Therefore, if a child does not chant in
consistent tempo but performs the correct rhythm patterns out of tempo, the child will still
receive a score of 1. Only a child who performs in consistent tempo will be rated based on
accuracy of rhythm patterns. However, if a judge tends to privilege rhythm over tempo or is less
strict on tempo, agreement among judges will suffer. Therefore, I made this clear in the
instructions for using the rating scale.

	
  

83	
  

Figure 11
The Original and Revised Rating Scales for Chant Performance Dimension
RHYTHM BEHAVIOR: CHANT PERFORMANCE
ORIGINAL
5. The child performs the entire chant with correct rhythmic patterns and consistent tempo.
4. The child performs most of the chant with correct rhythmic patterns and consistent tempo.
3. The child performs parts of the chant with correct rhythmic patterns and consistent tempo.
2. The child purposefully performs the chant; tempo is somewhat consistent but rhythmic
patterns are not accurate.
1. The child randomly performs the chant; tempo is not consistent and rhythmic patterns
are not accurate.
REVISED
5. The child performs the entire chant with consistent tempo and with correct rhythmic
patterns.
4. The child performs the chant with consistent tempo and more than half but not all of the
chant is performed with correct rhythmic patterns.
3. The child performs the chant with consistent tempo and some but less than half of the
chant is performed with correct rhythmic patterns.
2. The child performs the chant with consistent tempo but with incorrect rhythmic patterns.
1. The child performs the chant with inconsistent tempo.
Note. If the child does not perform with consistent tempo, the child should still receive
a score of 1 in spite of correct rhythmic patterns.
Rhythm Behavior: Rhythm Pattern Imitation
The original rating scale for Rhythm Pattern Imitation and its revision can be seen in
Figure 12. As with the previous dimension, the accuracy of Rhythm Pattern Imitation is
dependent upon both tempo and rhythmic accuracy. One of the judges asked, “What if the child
chants with accurate rhythm but not in tempo?” However, for the reasons described above, I
believe that both should be included in the dimension, as a pattern is not imitated accurately if it
is not imitated at the correct tempo. Moreover, lack of consistent tempo indicates that a child is
not feeling the macrobeat and microbeat, in which case the child cannot imitate rhythm patterns
correctly, as rhythm patterns are organized in relation to macro/microbeats. Because sense of
	
  

84	
  

macro/microbeat is essential in consistent tempo and consistent tempo is central to a child’s
rhythm development (Jordan, 1989), Rhythm Pattern Imitation dimension should include both. 	
  
The pilot study revealed that the children successfully responded to teacher-initiated
patterns, particularly in duple meter. The theoretical mean for the individual pattern imitation
dimensions was 3, and the observed means for patterns in duple meter were mostly around 4 and
were 3 for patterns in triple meter. This indicates that imitating duple rhythm patterns were
easier than imitating triple rhythm patterns for these children. However, there is no way to know
how to balance pattern difficulty between duple and triple rhythm patterns, so I was not able to
adjust this imbalance in the use of the rating scale.
Figure 12
The Original and Revised Rating Scales for Rhythm Pattern Imitation Dimension	
  
RHYTHM BEHAVIOR: RHYTHM PATTERN IMITATION
ORIGINAL
5. The child chants three patterns accurately in tempo.
4. The child chants two patterns accurately in tempo.
3. The child chants one pattern accurately in tempo.
2. The child chants the pattern in tempo but with inaccurate rhythms.
1. The child chants the pattern with inconsistent tempo and inaccurate rhythm.
REVISED
5. The child chants all patterns in tempo and with accurate rhythms.
4. The child chants two patterns in tempo and with accurate rhythms.
3. The child chants one pattern in tempo and with accurate rhythms.
2. The child chants all patterns with inconsistent tempo but with somewhat accurate
rhythms.
1. The child chants all patterns with inconsistent tempo and with inaccurate rhythms.

	
  

85	
  

Rhythm Behavior: Rhythm Improvisation
The original rating scale for Rhythm Improvisation and its revision can be seen in Figure
13. One of the concerns about this dimension was that the rating scale would be difficult for the
judges to use, because the wording could be confusing and, in some cases, the judges needed to
assess if a child’s improvisatory patterns corresponded to beat as well as to count the number of
same/different patterns a child performed. The wide range of interjudge reliabilities (r = .54 .96) could have been a reflection of this.
In response to the validation panel’s low level of agreement (54 percent) when
rearranging the criteria, and the wide range of reliabilities from the pilot study, a substantial
revision of this dimension was necessary. First, I revised the criteria of the dimension with
consistency and clarity in mind. Therefore, a child performing patterns that are different from
those of the teacher was described as a child performing “improvisatory patterns.” The next
inconsistent wording was “within the meter” and “corresponds with the beat.” In order to
provide consistency with what the rating scale is measuring and to facilitate judges’ ability to
stay focus on one variable when rating, “within the meter” was removed from the scale.
Wording describing the degree of achievement was not clear. As can be seen in Table
3.4, only one validation panel member (Panel Member 5) arranged the criteria in the anticipated
order. A “purposeful improvisatory pattern” was meant to describe when a child attempts to
perform an improvisatory pattern but is not completely successful. A child is aware that his or
her pattern should be different from what teacher has performed, but, because of various reasons,
that child might perform incomplete patterns or patterns that did not correspond to beat.
Therefore it was meant to represent a lower achievement level than performing an improvisatory
pattern with a limited vocabulary. However, it is possible that the wording “limited” was

	
  

86	
  

perceived as representing less achievement than “purposefully,” which has positive connotations.
These criteria might have been particularly confusing, as both criteria also included “imitate
patterns as part of improvisation.” As a result, I excluded the word “purposefully” in the
revision of the scale.
The descriptions, “a limited rhythmic vocabulary” in criterion 4 and “a variety of rhythms”
in criterion 5 needed more specificity. In order to clarify this, I revised the dimension so that “a
limited rhythmic vocabulary” was represented by one or two different rhythmic patterns, and “a
variety of rhythmic vocabulary” was defined as more than two different rhythmic patterns.
Figure 13
The Original and Revised Rating Scales for Rhythm Improvisation Dimension
RHYTHM BEHAVIOR: RHYTHM IMPROVISATION
ORIGINAL
5. The child performs improvisatory patterns within the meter and uses a variety of
rhythmic vocabulary.
4. The child performs improvisatory patterns within the meter but uses a limited
rhythmic vocabulary and may imitate a few patterns as a part of the improvisation.
3. The child purposefully improvises patterns that corresponds with beat and may
imitate a few patterns as part of the improvisation.
2. The child improvises patterns without corresponding with the beat and may imitate a
few patterns as a part of the improvisation.
1. The child imitates teacher’s patterns.

	
  

	
  

REVISED
5. The child performs all improvisatory patterns that correspond with the beat and uses a
rich variety of rhythmic vocabulary (more than two different patterns).
4. The child performs all improvisatory patterns that correspond with the beat but uses a
limited rhythmic vocabulary (one or two different patterns).
3. The child performs improvisatory patterns that correspond with the beat and may
imitate one or two teacher patterns.
2. The child performs improvisatory patterns without discernable rhythmic organization
and may imitate one or two teacher patterns.
1. The child does not perform improvisatory patterns but imitates teacher’s patterns.

87	
  

Rhythm Behavior: Keeping Beat
The original rating scale for Keeping Beat and its revision can be seen in Figure 14. In
my revision of this dimension, I again addressed wording that described non-observable behavior
(i.e., randomly, anticipate). Four judges from the pilot study commented, and the data revealed
that, while some judges rated a combination of macrobeats and microbeats as moving to steady
beat, others were confused about whether to count the combination as steady beat. Therefore, in
order to clarify the meaning of “steady beat,” I provided an annotation explaining that, while a
child can choose to perform steady beat either with macrobeat or microbeat, a combination of
both should not be considered as performing steady beat. Therefore, this rating scale specifically
measures a child’s ability to select one level of beat and keep it steady throughout the song.
Figure 14
The Original and Revised Rating Scales for Keeping Beat Dimension
RHYTHM BEHAVIOR: KEEPING BEAT
ORIGINAL
5. The child maintains a steady beat that corresponds with the beat of the song for all of
the song and is able to anticipate the beat.
4. The child maintains a steady beat that corresponds with the beat of the song for most
but not all of the song.
3. The child maintains a steady beat that corresponds with the beat of the song for some
but not all of the song.
2. The child maintains a steady beat but does not correspond with the beat of the song.
1. The child randomly performs a beat.

	
  

88	
  

Figure 14 (cont’d)
REVISED
5. The child maintains a steady beat that corresponds with the beat of the song for all of
the song.
4. The child maintains a steady beat that corresponds with the beat of the song for most
but not all of the song.
3. The child maintains a steady beat that corresponds with the beat of the song for less
than half but not all of the song.
2. The child maintains a steady beat that does not correspond with the beat of the song.
1. The child does not perform a steady beat.
Note. Steady beat refers to either macrobeat or microbeat. However, combination of both
will not be considered as “performing a steady beat” for this dimension.
Movement Behavior
The original rating scale for Movement Behavior and its revision can be seen in Figure
15. The main concern expressed about this dimension was that it measured too many aspects of
movement at once, which was confusing to the judges. The initial wording of the criteria within
the dimension sought to describe children’s use of Laban’s (1971) four movement elements
(flow, weight, time, space) in their movements. While this rating scale was constructed based on
a theoretical framework (Laban, 1971; Valerio et al., 1998), and early childhood music teachers
often model interaction of these elements in their movements, the pilot study revealed that this
movement dimension lacked reliability and therefore validity.
Comments from three of the pilot study judges informed me that, when children were
requested to “splash,” which was the image used to elicit flow with pulsation, most children
stopped flowing and focused on splashing. As a result, often times “splashing” looked more like
beat instead of flowing movement with pulsations. My inconsistent verbal instruction regarding
pulsations - “splash your fingers,” “splash around,” or sometimes no instruction - might have
exacerbated the problem, resulting in movement that did not exhibit flow with pulsation, even if

	
  

89	
  

the children had the capabilities to both flow and pulse at the same time. Moreover, verbal
directions such as, “Can you swim in the pool?” “Can you swim in the mud pool?” or “Can you
swim in the popcorn pool?” that were used to elicit different characteristics of Laban’s
movement elements failed to elicit the desired movement. Therefore, children’s movement was
difficult to assess using the original rating scale and assessment activities.
The comments of two pilot study judges aligned with those of two of the validation panel
members concerning the attempt to measure all four of Laban’s effort elements at once.
However, “it is difficult to experience flow, weight, time or space separately” (Jordan, 1989, p.
319). That is, when a teacher suggests children to swim in thick mud, the movement is
experienced by coordinating flow (bound), weight (heavy), time (sustained), and maybe space
(direct). There is a possibility that, because children did not have experiences in coordinating
these elements to the criterion song, they were confused when trying to exhibit the desired
movement qualities.
One validation panel member suggested including the coordination of breathing, moving,
and responding in this dimension, as coordination of these elements is an important part of
children’s movement development. However, I decided not to revise based upon this comment
for two reasons. First, capturing a child’s breathing would be difficult, because the camera
would need to be up-close, which would result in the inability to capture the movement of the
whole body. Second, attending to the coordination of these elements would be measuring a
different construct of movement behavior than was originally planned.
The content of the rating scale, children’s lack of experience in this particular movement
activity, and judges’ different understanding of movement elements are all possible factors that
jeopardized the reliability and the validity of this dimension of the scale. Therefore, in order to

	
  

90	
  

improve the rating scale, I substantially revised the scale to focus predominantly on flow, flow in
locomotion, and the coordination of flow and pulsation in locomotion. For the main study,
children explored movement with a criterion song, and they were not given any verbal
instruction during the activity.
Figure 15
The Original and Revised Rating Scales for Movement Behavior Dimension
MOVEMENT BEHAVIOR
ORIGINAL
5. The child coordinates flow, weight, space, and time in movement and is able to
pulsate to the beat, either spontaneously or upon request.
4. The child coordinates flow, weight, space, and time in movement.
3. The child purposely moves body in response to music and shows continuous flow.
2. The child randomly moves body, but the movement is not related to music.
1. The child does not move body while listening to music or singing/chanting.
REVISED
5. The child moves body in relation to music and uses flowing movements in locomotion
while also pulsing beat.
4. The child moves body in relation to music and uses flowing movements in
locomotion.
3. The child moves body in relation to music and uses flowing movements in place.
2. The child moves body, and the movement is related to music.
1. The child moves body, but the movement is not related to music.
Note. For example, elements related to music may include beat (i.e., walking, moving arms
or bending knees that correspond to beat).
Content validation is important in the development of a rating scale. The analysis of the
content validation process revealed the following overall concerns: (1) the validity of the criteria
in terms of content and sequence; (2) the validity of the construct that was being measured, (3)
clarity of the wording that described non-observable behaviors; (4) specificity of the wording
that described length or degree; and (5) consistency of wording to avoid confusion. The
comments of the judges were considered based on the intent of rating scale, assessment context,
	
  

91	
  

data collection procedures, its practical use in class, existing literature, and the researcher’s best
knowledge and experience as an early childhood music teacher and as the music teacher for the
participating children.

	
  

92	
  

CHAPTER VI
RESULTS
Reliabilities
Intrajudge Reliabilities
Since ECMBM was developed for use in the context of music instruction, it was essential
that I rate the performance during my actual teaching rather than only from recordings after class.
The extent to which the in-time ratings correlated with the ratings made while watching the
recordings would provide evidence that these scales could be used reliably in the classroom.
Therefore, I rated children’s musical behaviors during class while teaching and again while
watching or listening to the recorded data a week following each assessment to establish
intrajudge reliabilities.
The intrajudge reliabilities, means, standard deviations, and standard errors of
measurement of my ratings are presented in Table 4.1. The reliabilities all were high and ranged
from .84 to .96 for the individual dimensions. The observed means were relatively close to the
theoretical means except for those of the Duple Pattern Imitation and Keeping Beat: Gross Motor
dimensions, which were higher than the theoretical means. The observed standard deviations
were quite high when compared to the theoretical standard deviations for the children’s
performances in each dimension except for Duple and Triple Pattern Imitation, which means that
there was high variability among the scores and that the distributions were platykurtic rather than
normal.

	
  

93	
  

Table 4.1
Intrajudge Reliabilities, Means, Standard Deviations, and Standard Errors of Measurement for
Individual Dimensions
Intrajudge
r

First Rating
(TM=3; TSD=.67)
M
SD
SEM

Second Rating
(TM=3; TSD=.67)
M
SD
SEM

SA

.85

2.39

1.27

.21

2.92

1.25

.21

RT

.94

2.94

1.55

.26

2.89

1.43

.24

MJP

.92

2.81

1.41

.24

3.22

1.20

.20

MIP

.89

2.97

1.48

.25

3.17

1.32

.22

CH

.91

3.67

1.51

.25

3.53

1.42

.24

DP

.90

4.64

.59

.10

4.56

.65

.11

TP

.94

3.14

.83

.14

3.22

.83

.14

RI

.90

3.40

1.52

.25

3.06

1.43

.24

KBG

.91

4.03

1.32

.22

4.19

1.22

.20

KBM

.96

3.50

1.56

.26

3.58

1.46

.24

KBL

.96

3.33

1.60

.27

3.31

1.72

.29

MVT

.84

2.92

1.03

.17

2.92

1.03

.17

Note. (1) The score range for a single dimension is 1-5. (2) Abbreviations: SA = Singing
Accuracy; RT = Resting Tone; MJP = Major Pattern Imitation; MIP = Minor Pattern Imitation;
CH = Chant Performance; DP = Duple Pattern Imitation; TP = Triple Pattern Imitation; RI =
Rhythm Improvisation, KBG= Keeping Beat: Gross Motor; KBM = Keeping Beat:
Manipulatives; KBL = Keeping Beat: Locomotor; MVT = Movement
As can be seen in Table 4.2, the observed means and standard deviations of composite
scores in general were higher than the theoretical means and standard deviations. The standard
errors of measurement for each dimension and for the dimensions combined were small because
the reliabilities were high. This provides evidence that both sets of scores were meaningful;
none of the standard errors of measurement were greater than 1 for the individual dimensions of

	
  

94	
  

the ECMBM, and even the combined dimensions were smaller than 1. The only standard error
of measurement larger than 1 was for the entire measure, which had a score range of 12 to 60.
Table 4.2
Theoretical and Observed Means, Standard Deviations, and Standard Errors of Measurement
for Combined Dimensions
First Rating

Second Rating

TM
Tonal Behavior
(SA, RT, MJP, MIP)
Score Range: 4-20
Tonal Pattern
(MJP, MIP)
Score Range: 2-10
Rhythm Behavior
(CH, DP, TP, RI, KB)
Score Range: 7-35
Rhythm Pattern
(DP, TP)
Score Range: 2-10
Keeping Beat
(KBG, KBM, KBL)
Score Range: 3-15
ECMBM
(All dimensions)
Score Range: 12-60

TSD

M

SD

SEM

M

SD

SEM

12

2.67

11.11

4.73

.79

12.19

4.26

.71

6

1.33

5.78

2.67

.45

6.39

2.23

.37

21

4.67

25.69

5.42

.90

25.44

5.33

.89

6

1.33

7.78

1.29

.22

7.78

1.33

.22

9

2

10.86

3.56

.59

11.08

3.56

.59

36

8

39.72

9.43

1.57

40.56

8.81

1.47

Note. (1) See the note in Table 4.1 for dimension abbreviations. (2) TM = Theoretical Mean;
TSD = Theoretical Standard Deviation.
.

	
  

95	
  

Interjudge Reliabilities
Three independent judges rated the children’s performances, and I correlated these
ratings to determine interjudge reliabilities. All individual Tonal Behavior dimensions resulted
in high reliability coefficients ranging from .81 to .93 (see Table 5.1).
Table 5.1
Pearson Correlations for Individual Tonal Behavior Dimensions
Judge 1 and 2

Judge 1 and 3

Judge 2 and 3

Singing Accuracy

.86

.87

.93

Resting Tone

.90

.85

.83

Major Pattern Imitation

.85

.88

.81

Minor Pattern Imitation

.84

.88

.88

Table 5.2
Pearson Correlations for Individual Rhythm Behavior Dimensions
Judge 1 and 2

Judge 1 and 3

Judge 2 and 3

Chant Performance

.55

.55

.59

Duple Pattern Imitation

.91

.87

.80

Triple Pattern Imitation

.92

.91

.84

Rhythm Improvisation

.85

.82

.90

Keeping Beat: Gross Motor

.89

.88

.95

Keeping Beat: Manipulatives

.93

.86

.84

Keeping Beat: Locomotor

.89

.89

.88

In comparison to the overall high reliability coefficients reported in Tonal Behavior
dimensions, low reliability coefficients were found for the Chant Performance dimension of the
Rhythm Behavior dimensions. The other reliabilities within the Rhythm Behavior dimensions
were high. As can be seen in Table 5.2, the reliability coefficients for Chant Performance

	
  

96	
  

dimension were .55, .55, and .59. The reliability coefficients for rest of the dimensions ranged
from .80 to .95, all of which could be considered to be high for a single dimension.
The reliability coefficients for the Movement dimension were low to moderate. The
reliability coefficients were .53 between Judges 1 and 2; .54 between Judges 2 and 3; and .62
between Judges 1 and 3.
The reliability coefficients for the composite scores of dimensions are presented in Table
5.3. The interjudge reliabilities for the entire measure (ECMBM) were .94 and above, which are
high. The interjudge reliability coefficients for the composite scores of all Tonal Behavior
dimensions (i.e., Singing Accuracy, Resting Tone, and Major/Minor Pattern Imitation) were .94
and above. When Major and Minor Pattern Imitation dimensions were combined, the
reliabilities resulted a range of .90 - .94, which also is high.
Table 5.3
Pearson Correlations for Composite Scores of Combined Dimensions
Judge 1 and 2

Judge 1 and 3

Judge 2 and 3

Tonal Behavior Composite

.95

.94

.97

Tonal Pattern Composite

.90

.94

.92

Rhythm Behavior Composite

.90

.87

.91

Rhythm Pattern Composite

.94

.92

.88

Keeping Beat Composite

.95

.95

.95

ECMBM

.95

.94

.96

The reliability coefficients for the composite scores of the Rhythm Behavior dimensions
(i.e., Chant Performance, Duple/Triple Pattern Imitation, Rhythm Improvisation, and Keeping
Beat in Gross Motor, Manipulatives, and Locomotor modes) ranged from .87 to .91, all of which
are strong. When the ratings of Duple and Triple Pattern Imitation were combined, the
reliabilities resulted a range of .88 - .94. For the combined ratings of Keeping Beat dimensions
	
  

97	
  

of Gross Motor, Manipulatives, and Locomotor, the interjudge reliabilities all were .95, which
are high.
Means, Standard Deviations, and Standard Errors of Measurement of Final ECMBM
The means, standard deviations, and standard errors of measurement for each judge’s
ratings for individual dimensions were calculated and are reported below in Table 6.1.
Table 6.1
Means, Standard Deviations, and Standard Errors of Measurement of Individual Dimensions by
Judges
Judge 1

Judge 2

Judge 3

M

SD

SEM

M

SD

SEM

M

SD

SEM

SA

2.58

1.23

.21

2.97

1.23

.21

2.86

1.40

.23

RT

3.06

1.29

.21

3.06

1.43

.24

2.83

1.18

.20

MJP

3.14

1.22

.20

3.39

1.13

.19

2.89

1.24

.21

MIP

3.03

1.32

.22

2.94

1.26

.21

3.11

1.35

.23

CH

3.25

1.90

.32

3.03

1.52

.25

2.67

1.84

.31

DP

4.31

.86

.14

4.44

.81

.14

4.17

.81

.14

TP

3.25

.84

.14

3.39

.87

.15

3.17

.85

.14

RI

3.31

1.53

.26

3.06

1.37

.23

3.17

1.40

.23

KBG

4.28

1.06

.18

4.11

1.24

.21

4.22

1.12

.19

KBM

3.83

1.32

.22

3.75

1.50

.25

3.67

1.29

.22

KBL

3.78

1.27

.21

3.72

1.49

.25

3.42

1.63

.27

MVT

3.50

.74

.12

3.81

1.06

.18

3.08

1.03

.17

Note. (1) Abbreviations: SA = Singing Accuracy; RT = Resting Tone; MJP = Major Pattern
Imitation; MIP = Minor Pattern Imitation; CH = Chant Performance; DP = Duple Pattern
Imitation; TP = Triple Pattern Imitation; RI = Rhythm Improvisation, KBG= Keeping Beat:
Gross Motor; KBM = Keeping Beat: Manipulatives; KBL = Keeping Beat: Locomotor; MVT =
Movement (2) The score range is 1-5 for a single dimension.

	
  

98	
  

The theoretical mean for each single dimension was 3, and the theoretical standard
deviation was .67. As can be seen in Table 6.1, the observed means mostly approximated the
theoretical means except for the following dimensions: Duple Pattern Imitation, the Keeping
Beat dimensions (except for Judge 3 in the Locomotor mode), and for Judge 2 for the Movement
dimension. These observed means were quite a bit higher than the theoretical means, meaning
that the scales or the tasks associated with the assessments were too easy for the students. Many
of the observed standard deviations of judges were quite high in comparison to the theoretical
standard deviations, meaning that the scores were widely distributed and the distribution was
platykurtic or even possibly bi-modal in several dimensions.	
  
The observed means, and standard deviations of each judge’s composite scores are
presented in Table 6.2. The observed means for composite Tonal Behavior (i.e., all Tonal
Behavior dimensions combined) and Tonal Pattern Imitation (i.e., Major and Minor Pattern
Imitation dimensions combined) were fairly consistent with the theoretical means, whereas the
observed means for the Rhythm Behavior dimensions (i.e., all Rhythm Behavior dimensions
combined) were much higher than their theoretical counterparts. The observed standard
deviations were higher than the theoretical standard deviations for all Tonal Composite
dimensions and for the Keeping Beat dimensions.

	
  

99	
  

Table 6.2
Theoretical and Observed Means and Standard Deviations of Composite Scores by Judges
Judge 1

Judge 2

Judge 3

TM

TSD

M

SD

M

SD

M

SD

Tonal Behaviors
SA, RT, MJP, MIP
(Range = 4-20)

12

2.67

11.81

4.05

12.36

4.20

11.69

4.05

Tonal Pattern
MJP, MIP
(Range = 2-10)

6

1.33

6.17

2.26

6.33

2.17

6.00

2.27

21

4.67

26

4.99

25.50

5.13

24.47

4.69

6

1.33

7.56

1.48

7.83

1.52

7.33

1.47

11

2

11.89

2.87

11.58

3.26

11.31

3.14

36

8

41.31

8.50

41.67

8.72

39.25

8.10

Rhythm Behaviors
CH, DP, TP, RI, KB
(Range = 7-35)
Rhythm Pattern
DP, TP
(Range = 2-10)
Keeping Beat
KBG, KBM, KBL
(Range = 3-15)
ECMBM
All dimensions
(Range = 12-60)

Note. See the note in Table 6.1 for individual dimension abbreviations.
Intercorrelations of Dimensions
Intercorrelations among the 12 dimensions for the judges’ combined ratings are presented
in Table 7. Intercorrelations among Tonal Behavior dimensions resulted a range of .44 - .68.
The highest intercorrelation (.68) was reported between Singing Accuracy and Major Tonal
Pattern Imitation. This is not surprising, as both are tonal dimensions and require singing. I also
found moderate intercorrelations between Major and Minor Tonal Pattern Imitation (.66) as
would be expected.

	
  

100	
  

Table 7
Intercorrelations of the 12 Dimensions Using the Judges’ Combined Ratings of Individual
Dimensions
SA
RT
MJP
MIP
CH
DP
TP
RI
KBG
KBM
KBL

RT
.64

MJP
.68
.44

MIP
.61
.47
.66

CH
.33
.19
.16
.35

DP
.38
.29
.33
.41
.28

TP
.28
.34
.19
.48
.40
.63

RI
.38
.12
.33
.51
.42
.24
.23

KBG KBM KBL MVT
.24
.31
.36
.19
.14
.16
.28
.32
.33
.50
.53
.21
.34
.32
.43
.41
.00 -. 01 -.12
.02
.26
.16
.14
.17
.18
.33
.02
.13
.18
.08
.14
.06
.54
.41
.11
.40
.00
.25

Within Rhythm Behavior dimensions, two slightly negative intercorrelations were
reported between Chant Performance and Keeping Beat: Manipulatives (-.01) and Keeping Beat:
Locomotor (-.12) dimensions. These findings are surprising, as both are considered to be
rhythmic behaviors. The lowest correlation (-.12) was reported between Chant Performance and
Keeping Beat: Locomotor and the highest correlation (.63) was reported between the Duple
Pattern and Triple Pattern Imitation dimensions, which is not surprising as they measured the
same behavior with different content. Chant Performance had no meaningful correlation with
Keeping Beat: Gross Motor dimension. The Keeping Beat dimensions reported fairly consistent
and higher intercorrelations (.40 -.54) than the other dimensions within Rhythm Behavior
dimensions with the exception of the two pattern imitation dimensions (.63).
The results indicated that there were higher intercorrelations within dimensions than there
were between dimensions of different behaviors, which provide evidence of the validity of the
rating scale. And, the intercorrelations all were substantially lower than the reliabilities of the
dimensions, again, providing evidence of the validity of the measure. The intercorrelations of

	
  

101	
  

Tonal Behavior dimensions were within a range of .44 - .68, which is higher the correlations of
the Tonal Behavior dimensions to non-tonal dimensions, which is evidence of the validity of the
measure. These intercorrelations for the Tonal Behavior dimensions were within the range of .12
to .53 with Rhythm Behavior dimensions and the range of .19 to .41 with the Movement
Behavior dimension. The Movement Behavior dimension showed no or very low correlations
with all of the other dimensions (i.e., Tonal Behavior and Rhythm Behavior) as expected and
desired in terms of providing evidence of the validity of the dimension. However, it is possible
that the low reliability for this dimension attenuated its relationships with the other dimensions.
The ratings of individual dimensions within Rhythm Behaviors resulted in lower
intercorrelations (ranging from - .12 to .63) than intercorrelations within the Tonal Behavior
dimensions. The Chant Performance dimension showed no or slightly negative correlations with
Keeping Beat dimensions (Gross Motor, Manipulatives, Locomotor), which could be due to the
performance modes for each of the dimensions. As can be seen in Table 7, Chant Performance
showed higher correlations with dimensions that required vocal performance than with
dimensions that required movement (the Keeping Beat dimensions). The two dimensions (i.e.,
Chant Performance and Movement Behavior) with the low reliabilities had low intercorrelations
with other dimensions, as would be expected. It is difficult to find a consistent relationship with
a variable that is not consistent within itself. Overall, these results support the inverse validity of
the measure, with the exception of those dimensions that has low reliabilities.
Interpretations and Discussions
Comparison of the Results of the Pilot and Main Study
The interjudge reliabilities for the composite scores of Tonal Behavior dimensions in the
main study (r = .94 - .97) were improved from the pilot study (r = .83 - .96) as a result of the

	
  

102	
  

revisions to the measure. Whereas Tonal Pattern Imitation (Major and Minor Pattern) resulted in
a few moderate reliabilities in the pilot study, all of the reliabilities in the main study for these
dimensions were high.
The interjudge reliabilities of all Rhythm Behavior dimensions were improved in the
main study except for those of the Chant Performance dimension. The reported reliabilities of
the Chant Performance dimension in the pilot study were mostly moderate (n = 6) with one low
and a few high reliabilities (n = 3). In comparison, the reliabilities in the main study were low (r
= .55; .55; .59). The reliabilities for the Rhythm Pattern Imitation dimensions improved from
those of the pilot study. The reliability range for the combined ratings of the Rhythm Pattern
Imitation dimensions was .63 - .91 for the pilot study and was .88 - .94 for the main study. The
performance of the triple Pattern Imitation dimension was much more reliable in the main study
(r = .84 - .92) than in the pilot study (r = .20 - 1.00). The Rhythm Improvisation dimension also
showed improvement in its reliabilities. Although the pilot study reported a high reliability (r
= .96), the range was wider (r = .54 - .96) than the reliability range in the main study (r = .82 .90), which was consistently high.
The high reliabilities in the Keeping Beat dimensions indicated that the rating scales are
reliable for use when assessing children’s beat keeping performance in different performance
modes. With the revised assessment activities and my concentration on giving clear and
consistent performance directions, the Gross Motor (r = .88; .89; .95) and Locomotor (r
= .88; .89; .89) dimensions particularly resulted in higher and more consistent reliabilities than
during the pilot study (Gross motor r = .35 - .87; Locomotor r = 49 - .87). The result of the
Keeping Beat dimensions indicated that the clarification of the directions increased the
reliabilities to a great extent.

	
  

103	
  

The reliabilities of the Movement dimension did not improve from the pilot to the main
study. The Movement dimension reported reliability ranges of .20 - .85 in the pilot study and .53
- .62 in the main study. Even though the range of reliabilities was narrower in the main study
than in the pilot study, they were low reliabilities, so the dimension did not work well.
Overall, the revisions of the rating scales, the assessment activities, and the data
collection procedures, as well as the increased number of subjects improved reliabilities from the
pilot to the main study. However, the reliabilities for the Chant Performance and Movement
dimensions remained low.
Intrajudge Reliabilities
I did not rate all of the students during instruction in the pilot study, as I was still refining
data collection techniques. Therefore, there were no intrajudge reliabilities from the pilot study
that could be compared to those of the main study. In the main study, the intrajudge reliability
coefficients were high for all individual and combined dimensions. Only Singing Accuracy,
Minor Pattern Imitation, and the Movement dimensions had reliability coefficients less than .90
at .85, .89, and .84 respectively, while all other dimensions resulted in reliability coefficients
of .90 or above. The high intrajudge reliabilities indicate that all dimensions of ECMBM are
reliable for use in the context of music activities in an actual teaching environment. The ratings
that I did while teaching were highly related to my ratings of the audio/video recordings outside
of the class context. I used several techniques to facilitate my rating in class. I memorized the
criteria of the rating scales, in addition to referring to the rating card while I was assessing.
Moreover, I practiced the rating process several times with other preschool children prior to
actual assessment so that I felt comfortable leading the activities and rating simultaneously.
However, these techniques could be accomplished easily by a music teacher.

	
  

104	
  

Discussion
The interjudge reliabilities of the individual dimensions were mostly high. However, the
Chant Performance and Movement dimensions resulted in low and/or moderate reliabilities.
Rhythm Behavior: Chant Performance
The low reliability coefficients for the Chant Performance dimension were surprising, as
I had hoped that they would improve after revising the rating scales. Evidently, in spite of my
3

intention to improve reliability by including the note as a result of feedback from the judges in
the pilot study and the validation panel, this statement was not enough to elicit reliable ratings
from the judges. Perhaps the judges did not know how to interpret this direction or interpreted it
in different ways. If all judges had rated the children’s chant performances in the same way
based strictly on the instructions, there would have been lower means but higher interjudge
reliabilities, which was not the case.
The criterion chant also could have been a factor in the low reliability coefficients. One
of the four measures in the “Pancake” chant includes two eighth rests within each macrobeat (see
Appendix H). Rest patterns are difficult for children to audiate, as they need to internalize the
meter and microbeats to maintain a consistent tempo during rests. Some children rushed through
measure 3 because they were not audiating during the rests. When this happened, the judges
were forced to make subjective decisions about how to interpret this behavior. While one judge
may have tolerated rushing and given a score of 5, because all of the patterns were performed
correctly, another judge may have given the same performance a score of 1 because of my note
prioritizing a consistent tempo.
	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
3
	
  “If the child does not perform with consistent tempo, the child should still receive a score of 1
in spite of correct rhythmic patterns.”	
  
	
  
	
  

105	
  

Movement Behavior
As with the Chant Performance dimension, the Movement dimension resulted in low (r
= .53, .54) and moderate (r = .62) reliabilities. In spite of the substantive revision of the
dimension, the criteria of this rating scale did not result in high enough reliability coefficients to
support using the measure with children. The judges’ individual understanding of flow,
locomotor, and pulsation, or the observing and of rating these elements in coordination could
have been possible factors in the low reliabilities. In order to attain higher reliabilities, perhaps
there is a need to prepare judges more fully before rating so that they clearly understand how
these movement elements are exhibited. Since judges’ mutual understanding of criteria is critical
in rating reliably, a more detailed explanation of this rating scale as well as judge’s training may
have been helpful.
It is possible that an additive rather than a continuous rating scale is more appropriate for
rating these movement behaviors of children, especially if the children are not yet able to
coordinate these elements of movement. The Movement dimension was constructed based on
my observations as the subjects’ music teacher and on the research literature on movement
development, in addition to consideration of the curriculum of the school. I have observed many
children coordinating flow, locomotor movement, and pulsation when they, as a group, were
engaged in movement activities in class. However, without my moving with them, many
children were not able to exhibit coordination of these elements of movements. Perhaps the
influence of teacher modeling is a significant factor in eliciting the coordination of these
movement elements. However, individual movement elements (i.e., flow, locomotor movement,
pulsation) were identified easily. For example, many children moved in locomotor motion but
did not use true “flow.” Also, those children who demonstrated flow did not usually move in

	
  

106	
  

locomotion. Only a few children were able to use “flow” while moving through space, and
pulsation also was sparse. Until the children are able to coordinate the elements of movements
without teacher modeling and do them simultaneously, use of an additive rating scale or a
checklist may be more appropriate.
Intercorrelations
Mostly low intercorrelations between tonal, rhythm, and movement behaviors and higher
but not high intercorrelations within tonal and sometimes within rhythm behaviors provide
evidence that the dimensions of the scale do measure the different musical behaviors. The
moderate intercorrelations between the Major and Minor Pattern Imitation dimensions (.66) and
between the Duple and Triple Pattern Imitation dimensions (.63) were expected, as they not only
used the same rating scales, but also measure the same behavior, only with different content.
The same rating scale was used for the three modes of Keeping Beat, but low intercorrelations
were found between those dimensions (.40 - .54). Perhaps motor coordination played a role in
the student performances of some of those dimensions more than others, resulting in reliable but
unique dimensions. The ratings for the Chant Performance dimension resulted in slightly
negative or no intercorrelations with all keeping beat and movement dimensions. It is possible
that the low reliability coefficients of the Chant Performance dimension greatly attenuated the
intercorrelations. It also is possible that the vocal mode of performance of that dimension
contributed more to the scores than its rhythmic focus. Also, performing a chant successfully
required more focus on the surface rhythms of the chant than the underpinnings of beat and
meter, which were the focus of the Keeping Beat dimensions of the measure.
Although reliability contributes to validity, high reliability does not necessarily mean that
a measure is valid. However, intercorrelations and interjudge reliabilities do give evidence of the

	
  

107	
  

validity of a rating scale. This study suggests that the judges’ common understanding of the
criteria, the appropriateness of the assessment activities, and the use of appropriate criterion
songs/chants or patterns contributed to the reliabilities of those dimensions of the rating scale
with high reliabilities. The less successful Chant Performance dimension may have suffered
from a lack of clarity in how to use the rating scale as well as from the choice of a chant that
included rest patterns that were too difficult for the children. In order to improve reliabilities, the
Chant Performance dimension should be revised to include a detailed description of how to use
the scale, and the criterion chant should be carefully selected in order to avoid patterns that few
children can perform with a consistent tempo. Training of the judges on how to use the rating
scale also might result in improving their understanding of the scale and might increase the
reliabilities of both the Chant Performance and Movement dimensions.

	
  

108	
  

CHAPTER VII
SUMMARY, IMPLICATIONS,
RECOMMENDATIONS, AND CONCLUSION
Summary
Purpose and Research Questions
The purpose of this study was to develop a measure to assess the musical behaviors of
children in the context of regular music instruction using continuous rating scales, to establish
validity and reliability of the scales, and to develop appropriate assessment activities that can be
used as part of instruction. Specific research questions for the current study were as follows:
1. Can subjective validity be established for a researcher-designed instrument developed to
measure the musical behaviors of preschool children in the context of music instruction?
2. What are the intrajudge and interjudge reliabilities of a researcher-designed instrument
developed to measure the musical behaviors of preschool children in the context of
music instruction?
Design and Analysis
I designed the Early Childhood Musical Behavior Measure (ECMBM), which was
comprised of eight continuous rating scales, for use in the 12 dimensions of this study. In
addition to constructing the measure, I composed criterion songs and chants, selected tonal and
rhythm patterns, and designed assessment activities for use with the rating scale.
The subjects in this study were 4- and 5-year-old children enrolled in a Pre-K program at
an inner-city private school located in Detroit, Michigan. Children who were enrolled at this
particular school were mostly African-American, and 59 percent of the children in Pre-K
program receive reduced or free lunch. Thirty-six children in two intact classes participated in

	
  

109	
  

the study. Children had music class for 40 minutes, two times per week, and the instruction was
based on Music Learning Theory (Gordon, 2013). I served both as the researcher and as the
regular music teacher for the children who participated in the study.
I conducted a pilot study of the measure with 12 children who were not going to
participate in the main study. While I was conducting the pilot study, I sent the preliminary
ECMBM and assessment activities to seven validation panel members for content validation.
Each validation panel member rearranged the randomly ordered criteria in sequential order to
confirm that they represented a developmental sequence and provided detailed comments about
each of the rating scales and accompanying activities. Following the pilot study, I revised the
preliminary ECMBM and assessment activities based upon the comments of the judges from the
pilot study and the validation panel, my own experiences as the teacher and a judge, and the
results of the pilot study. Two validation panel members and my advisor reviewed the final
ECMBM before using it for the main study.
As I also was the children’s music teacher, I assessed the children’s musical behavior for
the main study during the children’s regular music classes. Activities from 12 classes over 6
weeks were video- and/or audio-recorded for subsequent rating. In addition to leading music
activities in the music classes, I taught each criterion song and chant to the children for 4 weeks
(eight classes) prior to the beginning of data collection.
A voice recorder was used to record children’s individual vocal performances and a video
camera was used to record the performances that required the judges to see the children’s
movement. Three independent judges who were experts in early childhood music rated the
children’s performance using ECMBM. In addition, I rated their performances both during class
and from the recordings at least one week after the data collection.

	
  

110	
  

I calculated interjudge reliability coefficients for each individual dimension and for the
dimensions combined. In addition to interjudge reliability, I correlated my ratings of the
children’s performances during class while collecting data and after class from listening to the
recorded data to obtain the intrajudge reliabilities and to determine whether the ratings that I
made during class were sufficiently reliable to allow the measure to be used during music
instruction by a music teacher. Means, standard deviations, and standard errors of measurement
were calculated for each of the judges’ ratings on individual and combined rating scales. Using
the combined ratings of three judges, I calculated the intercorrelations of the 12 dimensions to
serve as evidence of the inverse validity of the measure.
Results and Discussions
The results of content validation indicated that the validation panel members agreed 100
percent with the sequence of criteria for the Tonal Pattern Imitation, Rhythm Pattern Imitation,
and Keeping Beat rating scales; 94 percent for the Singing Accuracy, Resting Tone, and Chant
Performance rating scales; and 54 percent for the Rhythm Improvisation rating scale. While all
validation panel members validated both the content and the sequence of criteria for the
Movement rating scale, one member did not believe that the criteria were valid in terms of their
content and, therefore, did not provide responses. The comments and feedback from the four
independent judges who participated in the pilot study and the seven validation panel members
were considered when revising the ECMBM.
The intrajudge reliability coefficients of individual dimensions for Tonal Behaviors were
high, ranging from .85 to .94. The combined ratings of Major and Minor Pattern Imitation
dimensions (r = .94) and all combined Tonal Behavior dimensions (r = .95) resulted in high
intrajudge reliabilities as well. The individual Rhythm Behavior dimensions also resulted in high

	
  

111	
  

intrajudge reliabilities ranging from .90 to .96. Of all of the Rhythm Behavior dimensions,
Keeping Beat in Manipulatives and Locomotor modes resulted the highest intrajudge reliability
coefficients, which were .96 for both dimensions. The intrajudge reliabilities of the combined
ratings for Rhythm Pattern Imitation (Duple/Triple, r = .95), Keeping Beat (r = .98), and all of
the Rhythm Behaviors were high (r = .97). The reliability coefficient for the Movement
dimension was high (r = .84), but it was the lowest of all dimensions. As might be expected, the
highest intrajudge reliability (r = .98) was for the composite of ECMBM, which is strong
evidence of the reliability of the overall scale.
The interjudge reliability coefficients of individual dimensions for Tonal Behaviors were
high also ranging from .81 to .93. Singing Accuracy between Judges 2 and 3 and Resting Tone
between Judges 1 and 2 had reliabilities .90 and above. All other reliabilities ranged between .81
and .88. The interjudge reliability coefficients for individual Rhythm Behavior dimensions
ranged from .80 to .95 except for Chant Performance dimension (r = .55 - .59). The reliability
coefficients for the Movement dimension resulted a range of .53 - .62. The composite scores of
combined dimensions mostly were above .90 except for Rhythm Behavior Composite between
Judges 1 and 3, which was .87 and Rhythm Pattern Composite between Judges 2 and 3, which
was .88. The reliability coefficients for the composite of ECMBM ranged from .94 to .96.
The mean scores of individual Rhythm Behavior dimensions were fairly consistent with
the theoretical means of 3 except for Duple Pattern Imitation (M = 4.17 - 4.44), all three Keeping
Beat (M = 3.67 - 4.28) dimensions except for Judge 3 in Keeping Beat: Locomotor (M = 3.42),
and the Movement dimension for Judge 2 (M = 3.81). The standard deviations for the individual
dimensions were slightly higher (below 1.00) or much higher (above 1.00) than the theoretical
standard deviation of .67. The mean scores of the dimensions combined were close to the

	
  

112	
  

theoretical means except for all Rhythm Behavior dimensions combined for Judges 1 and 2 (TM
= 21; M = 26; 25.50), Rhythm Pattern Imitation dimensions combined for all judges (TM = 6; M
= 7.33 - 7.83), and Keeping Beat dimensions combined for Judge 1 (TM = 11; M = 11.89) and
Judge 2 (M = 11.58). While the ratings for the Tonal Behavior dimensions combined (SD = 4.05
- 4.20) and Tonal Pattern Imitation dimensions combined (SD = 2.17 - 2.27) resulted in much
higher standard deviations than their theoretical standard deviations, slightly higher standard
deviations were found for Rhythm Behavior dimensions combined (SD = 4.69 – 5.13), Rhythm
Pattern Imitation dimensions combined (SD = 1.47 -1.52), Keeping Beat dimensions combined
4

(SD = 2.87 - 3.26), and ECMBM as a whole (SD = 8.10 - 8.72) .
The results of the study further indicate that low intercorrelations among dimensions
were preponderant. However, a few moderate intercorrelations were reported between the Tonal
Behavior dimensions, except for between Resting Tone and each Tonal Pattern dimensions
(major/minor), and between the Duple and Triple Pattern dimensions. No relationships were
found between the dimensions of Rhythm Behavior and movement behavior, and two negative
intercorrelations were found between the Chant Performance dimension and Keeping Beat:
Manipulatives and Keeping Beat: Locomotor dimensions.
On the basis of the results of this study, it is reasonable to state that early childhood
music teachers who use a Music Learning Theory curriculum can use ECMBM to assist them in
assessing preschool children’s musical behaviors in the context of music instruction. The
intrajudge and interjudge reliabilities indicate that the Tonal Behavior rating scales are reliable,
and Rhythm Behavior rating scales also are reliable except for the Chant Performance dimension.
	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
4
Theoretical standard deviations: Tonal Behaviors = 2.67; Tonal Pattern Imitation combined
=1.33; Rhythm Behaviors = 4.67; Rhythm Pattern Imitation combined=1.33; all Keeping Beat
combined = 2; ECMBM = 8.
	
  

113	
  

The content validation process as well as the intercorrelations provides evidence that they also
are valid for use. The Movement dimension lacked content validity, and its ratings also resulted
in low to moderate reliabilities. These dimensions or the process of using them needs further
investigation and perhaps requires revision before they can be used with confidence in a
classroom.
Implications
Early childhood music teachers sometimes are hesitant to assess young children, because
they believe that their students are not yet developmentally ready for a structured assessment.
Moreover, assessment is time consuming, and children can be reluctant to perform on demand in
a structured setting. Yet, young children exhibit many musical behaviors as a result of music
instruction. A teacher’s systematic assessment of what each student can do musically could be
useful in planning instruction that facilitating the students’ musical development.
The current study offers a valid measurement tool that can be used effectively in the
context of music activities. Moreover, with 10 of the 12 dimensions of ECMBM, the teacher can
gather diagnostic information about the strengths of each child musically as well as about those
areas of their music learning that need more support and perhaps remediation. This information
could be used to improve a teacher’s ability to meet the specific instructional needs of each child.
Practically, several dimensions could be used in a single class period, and six to seven children
could be assessed during that class period on each dimension. In this way, children would be
engaged in several music activities that might occur during a normal instructional period,
preventing boredom, but the teacher could gather useful information for improving instruction.
Over several class periods, that teacher could gather information for every child on each of the
dimensions. Perhaps with the systematic documentation of each child’s musical growth over

	
  

114	
  

time using a valid measure like ECMBM, teachers can improve their instruction and provide
evidence of their students’ musical growth.
Developmentally appropriate assessment can serve as a connecting source for those (i.e.,
parents, caregivers, teacher, administrator, government) who are concerned with children’s
musical development. A developmentally appropriate and valid measurement not only helps to
define and support optimal music learning experiences for children, but the documentation of
children’s musical behaviors can serve as evidence of children’s musical growth to help teachers
communicate with administrators when seeking their support. Moreover, music teachers can use
this information when speaking with parents about their children’s musical development, why
music instruction is important, and how it should be provided. This could result in parents being
more enthusiastic about supporting their children’s music making at home and might guide them
to be more engaged with their children musically.
Recommendations for Future Research
These rating scales were validated with a specific set of activities and repertoire. Further
research is needed to determine whether a teacher can select other songs/chants or tonal/rhythm
patterns for use with ECMBM and customize their lessons throughout the year, so that children
can benefit from getting other musical experiences in the process of measuring their musical
skills and knowledge.
There is a need for further research that pertains to the refinement of the rating scales and
assessment activities. Additional changes, modifications or improvements are suggested below.
1.

The result of the Rhythm Pattern Imitation dimensions indicated that the criterion
duple patterns were too easy for the children. More difficult patterns should be
explored in conjunction with using the scale.

	
  

115	
  

2. Clearer directions or detailed notes should be added to the Chant Performance and
Movement dimensions to facilitate the judges’ rating process and to improve
reliability. In addition, a future study should provide training for the judges in terms
of how to use these dimensions of the scale.
3. Perhaps the Movement dimension should be revised as an additive rating scale or a
checklist rather than a continuous scale, as the movement behaviors did not seem to
be continuous in nature.
In addition, this study should be replicated with a larger sample, a different group of
children in terms of socio-economic status and racial make up, and with different criterion songs,
chants, or patterns to investigate further the validity of ECMBM. Perhaps similar scales should
be developed for older children as well, especially in light of the current culture of assessment in
the schools.
Conclusion
Children’s music learning can benefit from a teacher’s use of developmentally
appropriate, valid assessment tools, because with the information gained from such tools, a
teacher can individualize instruction based on an individual child’s musical achievements and
design activities to challenge high achieving students or to scaffold the learning of lower
achieving children. Consequently, music teachers should implement developmentally
appropriate assessments on a regular basis to provide optimal music learning experiences for
children. The current study was successful in validating much of the content of ECMBM with
music experts, and the interjudge reliabilities and intercorrelations of ECMBM provided
additional evidence of the validity of most of the measure. In addition, the data collection

	
  

116	
  

process was successful in measuring children’s musical behavior in the context of music
instruction.
Children should be immersed in music, and play with music during their preschool years.
In fact, playful music making and positive musical experiences should be the priority for
preschool children. However, music learning should be also based on and supported by
systematic curricula as well. Without teaching to satisfy children’s musical needs, children
could lose the “muse within” (Bjørkvold, 1992) as a result of engaging in dissatisfying music
learning experiences when they are young. The current study suggests that musical “play” for
children can provide rich data for teachers to collect and assess. When activities in music class
are playful, what children know and do musically is exhibited in its most natural form. Perhaps
through using measures such as ECMBM, assessment in the context of early childhood music
instruction can become something that early childhood music teachers do musically and
comfortably for and with children.

	
  

117	
  

APPENDICES

	
  

118	
  

Appendix A: Teacher Consent Form
Dear teachers,
My name is Gina J. Yi and I am a doctoral student in music education at Michigan State
University. I am conducting my Ph.D. dissertation research in fulfillment of the requirements for
the Doctor of Philosophy in Music Education degree at Michigan State University. Because
your students are the prospective participants of the study, and researchers are required to
provide a consent form to inform you about the study, I am writing to you to ask if I may
conduct my dissertation research with children who are in your class. You should feel free to ask
me any questions you may have.
Study Title: Development and Validation of a Musical Behavior Measure for Preschool Children
Researcher and Title: Gina J. Yi, Ph.D. candidate in Music Education at Michigan State
University, and Music Teacher at Cornerstone Schools
Department and Institution: College of Music, Michigan State University
1. PURPOSE OF RESEARCH:
For this dissertation, I will videotape preschool-aged children during their music class to
investigate the validity of the measure. The class will be videotaped for collecting data for rating
process; each class will be videotaped for 40 minutes, two times per week, for 6 weeks.
Videotaping will be conducted during participants’ regular music periods.
2. PRIVACY AND CONFIDENTIALITY:
At the conclusion of the study, videotapes will be locked in the office of the College of Music at
Michigan State University and destroyed after a period of three years. The data for this project
will be kept confidential. Student numbers will be used to help keep collected data anonymous.
Information about participants will be kept confidential to the maximum extent allowable by law.
The results of this study may be published or presented at professional meetings, but the
identities of all research participants will remain anonymous. The videotape will not be shown
in the public at any time.
	
  
3. CONTACT INFORMATION AND CONCERNS:
If you have any questions about this study or the videotaping process you may contact the
researcher:
Cynthia Taggart (primary investigator)
345 W. Circle Drive, Room 209
East Lansing, MI 48824
taggart@msu.edu
(517) 432-9678

	
  

Gina J. Yi (secondary investigator)
3030 Staten Ave. APT 5
Lansing, MI, 48910
ginayi@hotmail.com
(310) 991-0423

119	
  

If you have any questions or would like to register a complaint about this study, you may contact,
anonymously if you wish, the Michigan State University’s Human Research Protection Program
at 517-355-2180, FAX 517-432-4503, or e-mail irb@msu.edu or regular mail at 408 W. Circle
Dr., Room 207 Olds Hall, MSU, East Lansing, MI 48824.
Thank you for your time and consideration.
Sincerely,

Gina J. Yi
Ph.D. Candidate in Music Education at Michigan State University
Music Teacher at Cornerstone Schools
4. DOCUMENTATION OF INFORMED CONSENT:
Your signature below means that you are informed about this research study, and you allow the
research to be conducted in your students’ music class.
_______________________ ______________________________ ______________
Teacher’s Name
Teacher’s Signature
Date
You will be given a copy of this form to keep.

	
  

120	
  

Appendix B: Parent Consent Form
Your child is being asked to participate in a research project. Researchers are required to
provide a consent form to inform you about the study, to convey that participation is voluntary,
to explain risks and benefits of participation, and to empower you to make an informed decision.
You should feel free to ask the researchers any questions you may have.
Study Title: Development and Validation of a Musical Behavior Measure for Preschool Children
Researcher and Title: Gina J. Yi, Ph.D. Candidate in Music Education at Michigan State
University and Music Teacher at Cornerstone School
Department and Institution: College of Music, Michigan State University
1. PURPOSE OF RESEARCH:
Your child is being asked to participate in a research study, Development and Validation of a
Musical Behavior Measure for Preschool Children. This study is being conducted in fulfillment
of the requirements for Doctor of Philosophy in Music Education Degree at Michigan State
University. Your child has been selected as a possible participant in this study because your
child has been identified as a preschool-aged child enrolled at a Pre-K program. From this study,
the researcher hopes to construct musical behavior measure for preschool-aged children and
investigate its validity. The researcher will teach and videotape music class your child is
participating. Your child’s participation in this study will take about 40 minutes, two times per
week for 6 weeks. Whether you choose to allow your child to participate or not in this research
project, it will have no affect on your child's grade(s) in my class. In addition, no data will be
analyzed or reviewed until all final grades are posted for the semester.
2. WHAT YOU WILL DO:
Your child’s responsibility to the study is to participate in his/her regular music class.
3. POTENTIAL BENEFITS:
The potential benefits to your child for taking part in this study are that your child may enjoy
music activities in class.
4. POTENTIAL RISKS:
There is no expected discomfort associated with this videotaping procedure.
5. PRIVACY AND CONFIDENTIALITY:
The data for this project will be kept confidential, with no identifying information included.
Your child is guaranteed confidentiality for this study, except for the purpose of professional
supervision; in this case, your child’s identity will remain anonymous to the supervisor.
Information about your child will be kept confidential to the maximum extent allowable by law,
and will not be released unless there is a danger to your child or others. The data will be kept on
a password-protected personal laptop. At the conclusion of the study all videotapes and data will
be locked in the office at the College of Music at Michigan State University, and destroyed after
a period of three years. The results of this study may be published or presented at professional
meetings, but the identities of all research participants will remain anonymous.

	
  

121	
  

6. YOUR RIGHTS TO PARTICIPATE, SAY NO, OR WITHDRAW:
 Participation in this research project is completely voluntary. You have the right to say
no.
 You may change your mind at any time and withdraw, with no consequences.
 You may choose not to answer specific questions or to stop participating at any time.
 Choosing not to participate or withdrawing from this study will not make any difference
in the quality of any services your child may receive.
 You will be told of any significant findings that develop during the course of the study
that may influence your willingness to continue to participate in the research.
 If you agree to allow videotaping of the music class in which your child will be
participating, you have the right to change your mind about your child being taped, to
have stopped any of taping you did agree to at any time, and to have any tapes made of
your child not used in the study.
7. COSTS AND COMPENSATION FOR BEING IN THE STUDY:
Your child will not receive money or any other form of compensation for participating in this
study.
8. CONTACT INFORMATION FOR QUESTIONS AND CONCERNS:
If you have any questions about this study, such as scientific issues, how to do any part of it, or
to report an injury, please contact the researcher:
Cynthia Taggart (primary investigator)
345 W. Circle Drive, Room 209
East Lansing, MI 48824
taggart@msu.edu
(517) 432-9678

Gina J. Yi (secondary investigator)
3030 Staten Ave. APT 5
Lansing, MI, 48910
ginayi@hotmail.com
(310) 991-0423

If you have any questions about your child’s role and rights as a research participant, or would
like to register a complaint about this study, you may contact, anonymously if you wish, the
Michigan State University’s Human Research Protection Program at 517-355-2180, FAX 517432-4503, or e-mail irb@msu.edu or regular mail at 408 W. Circle Dr., Room 207 Olds Hall,
MSU, East Lansing, MI 48824.
9. DOCUMENTATION OF INFORMED CONSENT:
Your signature below means that you voluntarily agree for your child’s participation in this
research study.
________________________________________ _____________________________
Child’s Name
Classroom Teacher’s Name
________________________________________ _____________________________
Signature (Parent)
Date
* You will be given a copy of this form to keep.

	
  

122	
  

Appendix C: Child Assent Form
Please read the below to your child

Ms. Yi would like you to be in her study. For her study, she wants to videotape music class.
You and other friends are going to sing, chant, move, dance, and make music with various props
and instruments. Would you like to be in her study? Do you have any questions?
Is it okay for Ms. Yi to videotape you? Point the smiley face for YES or the frowning face for
NO.
Figure 16
Smiley Faces in Child Assent Form

_____________________________
Child’s Name

________________________________
Child’s Response

_____________________________
Class Teacher’s Name

_________________________________
Date

	
  

123	
  

Appendix D: Validation Panel Members
Suzanne L. Burton, Ph.D.
Professor of Music Education
University of Delaware
Denise Guilbault, Ph.D.
Associate Professor of Music Education
Rhode Island College
Lisa H. Koops, Ph.D.
Associate Professor of Music Education
Case Western Reserve University
Herbert D. Marshall, Ph.D.
Associate Professor of Music Education
Baldwin Wallace University, Conservatory of Music
Jill Reese, Ph.D.
Assistant Professor of Music Education
State University of New York at Fredonia
Joanne Rutkowski, Ph.D.
Professor of Music Education
The Pennsylvania State University
Wendy Valerio, Ph.D.
Associate Professor of Music
University of South Carolina

	
  

124	
  

Appendix E: Validation Forms
DIRECTIONS FOR VALIDATION: The criteria of each rating scale are NOT in a sequential
order. Please rearrange the criteria in a sequential order (5 represents the highest and 1 the
lowest level of achievement). You may cut and paste, or use the table provided (see below table).
Then, provide comments and feedback on the measure in terms of its valid use for preschool
children in the context of music activities.
Criteria
#
5
4
3
2
1

Alphabet
Character
E
C
A
D
B

TONAL BEHAVIOR: SINGING ACCURACY
A. The child sings parts of the song on the correct pitch level and melodic contour.
B. The child does not use singing voice or sings in a monotone; tonal context and
melodic contour are absent.
C. The child sings most of the song on the correct pitch level and melodic contour.
D. The child’s sings the entire song on the correct pitch level and melodic contour.
E. The child sings with some correct melodic contour but on the wrong pitch level.
Singing Accuracy
5
4
3
2
1
Comments:
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________

	
  

125	
  

TONAL BEHAVIOR: RESTING TONE
A. The child sings resting tone on the correct pitch.
B. The child does not use singing voice.
C. The child sings resting tone on somewhat correct pitch.
D. The child uses singing voice but is unable to find the correct pitch.
E. The child anticipates resting tone, sings resting tone on the correct pitch.
Note. A resting tone is the tonic of the song. Student sings the resting tone - tonic pitch of the
song - in response to teacher’s dominant pitch of the song.
Resting Tone
5
4
3
2
1
Comments:
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
TONAL BEHAVIOR: TONAL PATTERN IMITATION
A. The child sings the patterns with incorrect contour and incorrect pitches; sings in monotone.
B. The child sings three patterns accurately.
C. The child sings the pattern with correct melodic contour but with inaccurate pitches.
D. The child sings two patterns accurately.
E. The child sings one pattern accurately.
Tonal Pattern
Imitation
5
4
3
2
1
Comments:
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________

	
  

126	
  

Figure 17
Criterion Tonal Patterns for Tonal Pattern Imitation Dimensions

CRITERION TONAL PATTERNS

(Taggart, 1994)
RHYTHM BEHAVIOR: CHANT PERFORMANCE
A. The child purposefully performs the song/chant; tempo is somewhat consistent but rhythmic
patterns are not accurate.
B. The child performs most of the song with correct rhythmic patterns and consistent tempo.
C. The child performs the entire song with correct rhythmic patterns and consistent tempo.
D. The child performs parts of the song with correct rhythmic patterns and consistent tempo.
E. The child randomly performs the song/chant; tempo is not consistent and rhythmic patterns
are not accurate.
Chant Performance
5
4
3
2
1
Comments:
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________

	
  

127	
  

RHYTHM BEHAVIOR: RHYTHM PATTERN IMITATION
A. The child chants the pattern in tempo but with inaccurate rhythms.
B. The child chants one pattern accurately in tempo.
C. The child chants the pattern with inconsistent tempo and inaccurate rhythm.
D. The child chants three patterns accurately in tempo.
E. The child chants two patterns accurately in tempo.
Rhythm Pattern
Imitation
5
4
3
2
1
Comments:
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
Figure 18
Criterion Rhythm Patterns for Rhythm Pattern Imitation Dimensions

CRITERION RHYTHM PATTERNS

(Taggart, 1994)

	
  

128	
  

RHYTHM BEHAVIOR: RHYTHM IMPROVISATION
A. The child improvises patterns without corresponding with the beat and may imitate a few
patterns as a part of the improvisation.
B. The child imitates teacher’s patterns.
C. The child performs improvisatory patterns within the meter and uses a variety of rhythmic
vocabulary.
D. The child performs improvisatory patterns within the meter but uses a limited rhythmic
vocabulary and may imitate a few patterns as a part of the improvisation.
E. The child purposefully improvises patterns that corresponds with beat and may imitate a few
patterns as part of the improvisation.
Rhythm
Improvisation
5
4
3
2
1
Comments:
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
Figure 19
Teacher-Initiated Rhythm Patterns for Rhythm Improvisation Dimension
TEACHER-INITIATED RHYTHM PATTERNS

	
  

129	
  

RHYTHM BEHAVIOR: KEEPING BEAT
A. The child maintains a steady beat that corresponds with the beat of the song for all of the song
and is able to anticipate the beat.
B. The child randomly performs a beat.
C. The child maintains a steady beat that does not correspond with the beat of the song.
D. The child maintains a steady beat that corresponds with the beat of the song for most but not
all of the song.
E. The child maintains a steady beat that corresponds with the beat of the song for some but not
all of the song.
Keeping Beat
5
4
3
2
1
Comments:
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
MOVEMENT BEHAVIOR
A. The child randomly moves body, but the movement is not related to music.
B. The child coordinates flow, weight, space, and time in movement.
C. The child purposely moves body in response to music and shows continuous flow.
D. The child coordinates flow, weight, space, and time in movement and is able to pulsate to the
beat, either spontaneously or upon request.
E. The child does not move body while listening to music or singing/chanting.
Movement
5
4
3
2
1
Comments:
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________

	
  

130	
  

Unfamiliar Words:
•
•
•
•
•
•

Pulsation refers to coordinating pulse or bumps while moving. The “pulse” can be shown as
macrobeat/microbeat of the song or chant.
Continuous flow is free, flexible, smooth movement that is uninterrupted. Moving one’s
whole body with continuous flow requires moving the hips, back, shoulders in circular
pathways and bending the knees.
Space can be stationary, locomotor, high, medium, or low. Also, movements can be
performed forward, backward, diagonally or sideways using space.
Time describes movement initially as being sustained (slow) or quick.
Weight describes the force behind a movement, such as strong or gentle.
Continuous flow with pulsation is free and flexible movement that, while uninterrupted as
in continuous flow, contains visual and physical representations of a consistent pulse.
(Valerio et al., 1998)

Figure 20
Movement Element Chart
Verbal Descriptions
Swim in the pool filled
with water
Swim in the pool filled
with mud
Swim in the pool filled
with popcorn
Swim and splash water
with your fingers

	
  

Flow

Weight
Moderately
Heavy

Space

Heavy
Continuous

Light

Light

131	
  

Time
Moderately
Slow

Pulsation

Slow
Low, high,
Locomotor

Quick

Quick

Shake
hands or
flick
fingers on
beat

Appendix F: Instructions and Rating Sheets
Judge Name:
INSTRUCTIONS:
•
•
•
•
•

Listen for one dimension ONLY before going to the next dimension. For example, listen to
all of the children’s performances of SINGING ACCURACY dimension, then, rate the next
dimension.
Carefully read all 5 criteria before rating. Please note that validity might be in jeopardy
when a single criterion is attended to without first considering all criteria in the dimension
(Gordon, 2002).
5 represents the highest and 1 the lowest level of achievement.
You may review the recordings (video/audio) as many times as necessary to form your rating.
You may choose to listen again to the performance of a child if you would like to reconsider
your rating.
5

COMMENTS :
Please provide comments or feedback concerning the rating process.
• Did you find any problem with the quality of audio/video data while rating?
• Did you have questions or confusion regarding the rating scale?
• Do you have feedback about my assessment instruction for the students?
• Other concerns or questions?

RATING SHEETS:
TONAL BEHAVIOR: SINGING ACCURACY
5. The child sings the entire song with correct melodic contour and on the correct pitches
with good intonation.
4.	
  The child sings the song with correct melodic contour but with a few incorrect pitches
or poor intonation.
3. The child sings the song with correct melodic contour but on the wrong pitches.
2. The child sings the song with some correct melodic contour but on the wrong pitches.
1. The child sings with a lack of tonal context and melodic contour is absent.

	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
5
This was only provided for the pilot study.
	
  

132	
  

Figure 21
A Rating Sheet for Judges to Record Children’s Singing Accuracy Scores
Child #
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19

SINGING ACCURACY
Score
Child #
W1
W2
W3
W4
W5
W6
W7
W8
W9
W10
W11
W12
W13
W14
W15
W16
W17

Figure 22
Criterion Song for Singing Accuracy Dimension

Autumn Leaves
Criterion Song

	
  

133	
  

Score

TONAL BEHAVIOR: RESTING TONE
DIRECTION: Listen to the audio file first and give a score for each child. Then, watch the
video for the child/children who you gave score 4 to determine score 4 or 5.
5. The child gives a preparatory breath in anticipation of the resting tone and sings resting tone
on the correct pitch.
4. The child sings resting tone on the correct pitch.
3. The child sings resting tone on somewhat correct pitch; slides into or adjusts to correct pitch.
2. The child is unable to find the correct resting tone pitch but sings within a whole step.
1. The child is unable to find the correct resting tone pitch.
Figure 23
A Rating Sheet for Judges to Record Children’s Resting Tone Scores
Child #
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19

	
  

RESTING TONE
Score
Child #
W1
W2
W3
W4
W5
W6
W7
W8
W9
W10
W11
W12
W13
W14
W15
W16
W17

134	
  

Score

TONAL BEHAVIOR: TONAL PATTERN IMITATION
5. The child sings all patterns accurately.
4. The child sings two patterns accurately.
3. The child sings one pattern accurately.
2. The child sings all patterns with correct melodic contour but with inaccurate pitches.
1. The child sings all patterns with incorrect melodic contour and inaccurate pitches.
Figure 24
Criterion Tonal Patterns for Tonal Pattern Imitation Dimensions

Figure 25
A Rating Sheet for Judges to Record Children’s Tonal Pattern Imitation Scores
Child #
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19
	
  

Major

TONAL PATTERN IMITATION
Minor
Child #
W1
W2
W3
W4
W5
W6
W7
W8
W9
W10
W11
W12
W13
W14
W15
W16
W17

135	
  

Major

Minor

RHYTHM BEHAVIOR: CHANT PERFORMANCE
5. The child performs the entire chant with consistent tempo and with correct rhythmic
patterns.
4. The child performs the chant with consistent tempo and more than half but not all of
the chant is performed with correct rhythmic patterns.
3. The child performs the chant with consistent tempo and some but less than half of the
chant is performed with correct rhythmic patterns.
2. The child performs the chant with consistent tempo but with incorrect rhythmic
patterns.
1. The child performs the chant with inconsistent tempo.
NOTE: If the child does not perform with consistent tempo, the child should still receive a score
of 1 in spite of correct rhythmic patterns.
Figure 26
A Rating Sheet for Judges to Record Children’s Chant Performance Scores
Child #
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19

	
  

CHANT PERFORMANCE
Score
Child #
W1
W2
W3
W4
W5
W6
W7
W8
W9
W10
W11
W12
W13
W14
W15
W16
W17

136	
  

Score

Figure 27
Criterion Chant for Chant Performance Dimension

Pancake
Criterion Chant

RHYTHM BEHAVIOR: RHYTHM PATTERN IMITATION
5. The child chants all patterns in tempo and with accurate rhythms.
4. The child chants two patterns in tempo and with accurate rhythms.
3. The child chants one pattern in tempo and with accurate rhythms.
2. The child chants all patterns with inconsistent tempo but with somewhat accurate rhythms.
1. The child chants all patterns with inconsistent tempo and with inaccurate rhythms.
Figure 28
A Rating Sheet for Judges to Record Children’s Rhythm Pattern Imitation Scores
Child #
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19

	
  

RHYTHM PATTERN IMITATION
Duple
Triple
Child #
Duple
W1
W2
W3
W4
W5
W6
W7
W8
W9
W10
W11
W12
W13
W14
W15
W16
W17

137	
  

Triple

Figure 29
Criterion Rhythm Patterns for Rhythm Pattern Imitation Dimensions

RHYTHM BEHAVIOR: RHYTHM IMPROVISATION
5. The child performs all improvisatory patterns that correspond with the beat and uses a rich
variety of rhythmic vocabulary (more than two different patterns).
4. The child performs all improvisatory patterns that correspond with the beat but uses a limited
rhythmic vocabulary (one or two different patterns).
3. The child performs improvisatory patterns that correspond with the beat and may imitate one
or two teacher patterns.
2. The child performs improvisatory patterns without discernable rhythmic organization and may
imitate one or two teacher patterns.
1. The child does not perform improvisatory patterns but imitates teacher’s patterns.
Figure 30
Teacher-Initiated Rhythm Patterns for Rhythm Improvisation Dimension

	
  

138	
  

Figure 31
A Rating Sheet for Judges to Record Children’s Rhythm Improvisation Scores
Child #
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19

RHYTHM IMPROVISATION
Score
Child #
W1
W2
W3
W4
W5
W6
W7
W8
W9
W10
W11
W12
W13
W14
W15
W16
W17

Score

RHYTHM BEHAVIOR: KEEPING BEAT
5. The child maintains a steady beat that corresponds with the beat of the song for all of
the song.
4. The child maintains a steady beat that corresponds with the beat of the song for most but not
all of the song.
3. The child maintains a steady beat that corresponds with the beat of the song for less than half
but not all of the song.
2. The child maintains a steady beat that does not correspond with the beat of the song.
1. The child does not perform a steady beat.
NOTE: Steady beat refers to either macrobeat or microbeat. However, combination of both will
not be considered as “performing a steady beat” for this dimension.

	
  

139	
  

Figure 32
A Rating Sheet for Judges to Record Children’s Keeping Beat Scores
Child #
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19

KEEPING BEAT: (MODE)
Score
Child #
W1
W2
W3
W4
W5
W6
W7
W8
W9
W10
W11
W12
W13
W14
W15
W16
W17

Score

MOVEMENT BEHAVIOR
5. The child moves body in relation to music and uses flowing movements in locomotion
while also pulsing beat.
4. The child moves body in relation to music and uses flowing movements in locomotion.
3. The child moves body in relation to music and uses flowing movements in place.
2. The child moves body, and the movement is related to music.
1. The child moves body, but the movement is not related to music.
NOTE: For example, elements related to music may include beat (i.e., walking, moving arms or
bending knees that correspond to beat).

	
  

140	
  

Figure 33
A Rating Sheet for Judges to Record Children’s Movement Scores
MOVEMENT
Child #
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
P13
P14
P15
P16
P17
P18
P19

	
  

Score

Child #
W1
W2
W3
W4
W5
W6
W7
W8
W9
W10
W11
W12
W13
W14
W15
W16
W17

141	
  

Score

Appendix G: Rating Cards
Figure 34
A Rating Card Used for Recording Children’s Singing Accuracy Scores
Child #

#1 - 6

#7 - 12

#13 - 17

Assessment Date

Group:
No

S

No

1

5. The child sings the entire song with correct melodic contour and on
the correct pitches with good intonation.
4.	
  The child sings the song with correct melodic contour but with a
few incorrect pitches or poor intonation.
3. The child sings the song with correct melodic contour but on the
wrong pitches.
2. The child sings the song with some correct melodic contour but on
the wrong pitches.
1. The child sings with a lack of tonal context, and melodic contour
is absent.

10

2

TONAL BEHAVIOR: SINGING ACCURACY

11

3

12

4

13

5

14

6

15

7

16

8

S

17

9

Figure 35
A Rating Card Used for Recording Children’s Resting Tone Scores
Child #

#1 - 6

#7 - 12

#13 - 17

Assessment Date

No

S

No

1

5. The child gives a preparatory breath in anticipation of the resting
tone and sings resting tone on the correct pitch.
4. The child sings resting tone on the correct pitch.
3. The child sings resting tone on somewhat correct pitch; slides into
or adjusts to correct pitch.
2. The child is unable to find the correct resting tone pitch but sings
within a whole step.
1. The child is unable to find the correct resting tone pitch.
	
  

142	
  

10

2

TONAL BEHVAIOR: RESTING TONE

	
  

Group:

11

3

12

4

13

5

14

6

15

7

16

8

17

9

S

Figure 36
A Rating Card Used for Recording Children’s Tonal Pattern Imitation Scores
Child #

#1 - 6

#7 - 12

#13 - 17

Assessment Date

Group:
No

S

No

5. The child sings all patterns accurately.
4. The child sings two patterns accurately.
3. The child sings one pattern accurately.
2. The child sings all patterns with correct melodic contour but with
inaccurate pitches.
1. The child sings all patterns with incorrect melodic contour and
inaccurate pitches.

1

10

2

11

3

12

4

13

5

14

6

15

7

16

8

TONAL BEHVAIOR: TONAL PATTERN IMITATION

17

9

	
  

143	
  

S

Figure 37
A Rating Card Used for Recording Children’s Chant Performance Scores
Child #

#1 - 6

#7 - 12

#13 - 17

Assessment Date

No

RHYTHM BEHAVIOR: CHANT PERFORMANCE
5. The child performs the entire chant with consistent tempo and with
correct rhythmic patterns.
4. The child performs the chant with consistent tempo and more than
half but not all of the chant is performed with correct rhythmic
patterns.
3. The child performs the chant with consistent tempo and some but
less than half of the chant is performed with correct rhythmic patterns.
2. The child performs the chant with consistent tempo but with
incorrect rhythmic patterns.
1. The child performs the chant with inconsistent tempo.
	
  

	
  

Group:

144	
  

S

No

1

11

2

12

3

13

4

14

5

15

6

16

7

17

8
9

S

Figure 38
A Rating Card Used for Recording Children’s Rhythm Pattern Imitation Scores
Child #

#1 - 6

#7 - 12

#13 - 17

Assessment Date

Group:
No

S

No

5. The child chants all patterns in tempo and with accurate rhythms.
4. The child chants two patterns in tempo and with accurate rhythms.
3. The child chants one pattern in tempo and with accurate rhythms.
2. The child chants all patterns with inconsistent tempo but with
somewhat accurate rhythms.
1. The child chants all patterns with inconsistent tempo and with
inaccurate rhythms.

1

10

2

11

3

12

4

13

5

14

6

15

7

16

8

RHYTHM BEHAIOVR: RHYTHM PATTERN IMITATION

17

9

	
  

145	
  

S

Figure 39
A Rating Card Used for Recording Children’s Rhythm Improvisation Scores
Child #

#1 - 6

#7 - 12

#13 - 17

Assessment Date

Group:
No

S

No

1

5. The child performs all improvisatory patterns that correspond with
the beat and uses a rich variety of rhythmic vocabulary (more than
two different patterns).
4. The child performs all improvisatory patterns that correspond with
the beat but uses a limited rhythmic vocabulary (one or two
different patterns).
3. The child performs improvisatory patterns that correspond with the
beat and may imitate one or two teacher patterns.
2. The child performs improvisatory patterns without discernable
rhythmic organization and may imitate one or two teacher patterns.
1. The child does not perform improvisatory patterns but imitates
teacher’s patterns.

10

2

RHYTHM BEHAVIOR: RHYTHM IMPROVISATION

11

3

12

4

13

5

14

6

15

7

16

8

S

17

9

Figure 40
A Rating Card Used for Recording Children’s Keeping Beat Scores
Child #

#1 - 6

#7 - 12

#13 - 17

Assessment Date

No

RHYTHM BEHAVIOR: KEEPING BEAT
5. The child maintains a steady beat that corresponds with the beat of
the song for all of the song.
4. The child maintains a steady beat that corresponds with the beat of
the song for most but not all of the song.
3. The child maintains a steady beat that corresponds with the beat of
the song for less than half but not all of the song.
2. The child maintains a steady beat that does not correspond with the
beat of the song.
1. The child does not perform a steady beat.	
  

	
  

Group:

146	
  

S

No

1

10

2

11

3

12

4

13

5

14

6

15

7

16

8

17

9

S

Figure 41
A Rating Card Used for Recording Children’s Movement Scores
Child #

#1 - 6

#7 - 12

#13 - 17

Assessment Date

Group:
No

No

1

5. The child moves body in relation to music and uses flowing
movements in locomotion while also pulsing beat.
4. The child moves body in relation to music and uses flowing
movements in locomotion.
3. The child moves body in relation to music and uses flowing
movements in place.
2. The child moves body, and the movement is related to music.
1. The child moves body, but the movement is not related to music	
  

10

2

MOVEMENT BEHAVIOR

11

3

12

4

13

5

14

6

15

7

16

8

17

9

	
  

S

147	
  

S

Appendix H: Criterion Songs and Chant
Figure 42
Criterion Songs and Chant Used for Assessment Activities

Autumn Leaves
Gina J. Yi

Love Somebody
American

	
  

148	
  

Figure 42 (cont’d)

Drumming
Gina J. Yi

Pancake
Gina J. Yi

	
  

149	
  

Figure 42 (cont’d)

Chirping Bird
Gina J. Yi

	
  

150	
  

REFERENCES

	
  

151	
  

REFERENCES

Abeles, H. F., Hoffer, C. R., & Klotman, R. H. (1994). Foundations of music education (2nd ed.).
New York, NY: Schirmer Books.
Alvarez, B. J. (1981). Preschool music education and research on the musical development of
preschool children: 1990 to 1980 (Doctoral dissertation). Retrieved from ProQuest
Dissertations and Theses database. (Order No. 8204585)
Andress, B. (1973). Music in early childhood. Washington: Music Educators National
Conference.
Andress, B. (1980). Music experiences in early childhood. New York: Holt, Rinehart and
Winston.
	
  
Aronoff, F. W. (1969). Music and young children. New York: Holt, Rinehart and Winston.
Bagnato, S. J. (2007). Authentic assessment for early childhood intervention: Best practices.
New York: The Guilford Press.
Barrett, M. (1997). Invented notations: A view of young children's musical thinking. Research
Studies in Music Education, 8, 2-14. doi: 10.1177/1321103X9700800102
Barrett, M. S. (2006). Inventing songs, inventing worlds: The 'genesis' of creative thought and
activity in young children's lives. International Journal of Early Years Education, 14(3),
201-220.
Beaty, J. J. (2010). Observing development of the young child (7th ed.). Upper Saddle River, NJ:
Merrill.
Bentzen, W. R. (2005). Seeing young children: A guide to observing and recording behavior.
Clifton Park, NY: Thomson/Delmar Learning.
Berger, A. A., & Cooper, S. (2003). Musical play: A case study of preschool children and
parents. Journal of Research in Music Education, 51(2), 151-165.
Bjørkvold, J. R. (1992). The muse within: Creativity and communication, song and play from
childhood through maturity (W. H. Halverson, Trans.). New York, NY: HarperCollins
Publishers.
Blesedell, D. S. (1991). A study of the effects of two types of movement instruction on the rhythm
achievement and developmental rhythm aptitude of preschool children (Doctoral
dissertation). Retrieved from ProQuest Dissertations and Theses database. (Order No.
9134919)

	
  

152	
  

Boardman, E., & Andress, B. (1981). The music book: Teacher's reference book, Grade K. New
York: Holt, Rinehart & Winston.
Boyle, J. D. (1989). Perspective on evaluation. Music Educators Journal, 76(4), 22-25.
Boyle, J. D. (1996). The national standards: Some implications for assessment. In Aiming for
excellence: The impact of the standards movement on music education (pp. 109-115).
Reston: VA: Music Educators National Conference.
Bruner, J. S. (1966). Toward a theory of instruction. Cambridge, MA: Belknap Press of Harvard
University.
Burton, S. (2002). An exploration of preschool children's spontaneous songs and chants. Visions
of Research in Music Education, 2, 7-16. Retrieved from http://users.rider.edu/~vrme/
Campbell, P. S. (1998). The musical cultures of children. Research Studies in Music Education,
11, 42-51. doi: 10.1177/1321103X9801100105
Campbell, P. S., & Scott-Kassner, C. (2006). Music in childhood: From preschool through the
elementary grades. Belmont, CA: Thomson Schirmer.
Copple, C., & Bredekamp, S. (Eds.). (2009). Developmentally appropriate practice in
early childhood programs (3rd ed.). Washington, DC: National Association for the
Education of Young Children.
Edwards, L. C. (2013). Music and movement: A way of life for the young child (7th ed.). Boston:
Pearson.
Etopio, E. A. (2009). Characteristics of early musical environments associated with preschool
children's music skills (Doctoral dissertation). Retrieved from ProQuest Dissertations and
Theses database. (UMI No. 3342135)
Etopio, E. A., & Cissoko, K. M. (2005). The hard work of music play: Establishing an
appropriate early childhood music environment. In M. Runfola & C. C. Taggart (Eds.),
The development and practical application of music learning theory (pp. 53-67). Chicago,
IL: GIA Publications.
Findlay, E. (1971). Rhythm and movement: Applications of Dalcroze eurhythmics.
Evanston, IL: Summy-Birchard Company.
Fiore, L. B. (2012). Assessment of young children: A collaborative approach. New York, NY:
Routledge.
Flohr, J. W. (1984). Young children's improvisations: A longitudinal study. Paper presented at
the Music Educators 49th National In-Service Conference, Chicago, IL. Retrieved from
ERIC database. (ED255318)

	
  

153	
  

Flohr, J. W., & Trollinger, V. L. (2010). Music in elementary education. Upper Saddle
River: Prentice Hall.
Forum on Educational Accountability [FEA]. (2007). Assessment and accountability for
improving schools and learning: Principles and recommendations for federal law and
state and local systems. Retrived from
www.edaccountability.org/AssessmentFullReportJUNE07.pdf
Fox, D. B. (1989). MusicTIME and music times two: The Eastman infant-toddler music
programs. In B. Andress (Ed.), Promising practices: Prekindergarten music education
(pp. 13-23). Reston, VA: Music Educators National Conference.
Gembris, H. (2002). The development of musical abilities. In R. Colwell & C. Richardson (Eds.),
The new handbook of research on music teaching and learning: A project of the Music
Educators National Conference (pp. 487-508). New York, NY: Oxford University Press.
Goetze, M. (1985). Factors affecting accuracy in children's singing (Doctoral dissertation).
Retrieved from ProQuest Dissertations and Theses database. (Order No. 8528488)
Gordon, E. E. (1976). Tonal and rhythm patterns: An objective analysis. Albany, NY: State
University of New York Press.
Gordon, E. E. (1989). Audie: A game for understanding and analyzing your child's music
potential. Chicago, IL: GIA Publications.
Gordon, E. E. (2002). Rating scales and their uses for measuring and evaluating achievement in
music performance. Chicago, IL: GIA Publications.
Gordon, E. E. (2012). Learning sequences in music: A contemporary music learning theory.
Chicago, IL: GIA Publications.
Gordon, E. E. (2013). A music learning theory for newborn and young children. Chicago, IL:
GIA Publications.
Greenberg, M. (1976). Research in music in early childhood education: A survey with
recommendations. Bulletin of the Council for Research in Music Education, 45, 1-20.
Guddemi, M. P. (2003). Assessments for children ages 3 to 8 years. In J. E. Wall & G. R. Walz
(Eds.), Measuring up: Assessment issues for teachers, counselors, and administrators (pp.
273-281). Greensboro, NC: ERIC Counseling and Student Services Clearinghouse.
Retrieved from ERIC database. (ED480379)
Hicks, W. K. (1993). An investigation of the initial stages of preparatory audiation
(Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database.
(Order No. 9316493)

	
  

154	
  

Hornbach, C. M. (2005). Ah-eee-ah-eee-yah-eee, bum, and pop, pop, pop: Teacher initiatives,
teacher silence, and children's vocal responses in early childhood music classes
(Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database.
(UMI No. 3189669)
Hornbach, C. M., & Taggart, C. C. (2005). The relationship between developmental tonal
aptitude and singing achievement among kindergarten, first-, second-, and third-grade
students. Journal of Research in Music Education, 53(4), 322-331.
Jacobi-Karna, K. L. (1996). The effects of the inclusion of text on the singing accuracy of
preschool children (Doctoral dissertation). Retrieved from ProQuest Dissertations and
Theses database. (UMI No. 9713419)
Jaques-Dalcroze, E. (1967). Rhythm, music and education. (H. F. Rubenstein Trans.). London:
Dalcroze Society.
Jersild, A. T., & Bienstock, S. F. (1934). A study of the development of children's ability to sing.
The Journal of Educational Psychology, XXV(7), 481-503.
Jersild, A. T., & Bienstock, S. F. (1935). Development of rhythm in young children. New York
City: Teachers College, Columbia University.
Jordan, J. (1989). Laban movement theory and how it can be used with music learning theory. In
D. L. Walters & C. C. Taggart (Eds.), Readings in music learning theory (pp. 316-333).
Chicago: GIA Publications.
Kierstead, J. K. (2006). Listening to the spontaneous music-making of preschool children in
play: Living a pedagogy of wonder (Doctoral dissertation). Retrieved from ProQuest
Dissertations and Theses database. (UMI No. 3241541)
Laban, R. (1971). The mastery of movement. London: MacDonald and Evans.
Lecanuet, J.-P. (1996). Prenatal auditory experience. In I. Deliège & J. A. Sloboda (Eds.),
Musical beginnings: Origins and development of musical competence (pp. 3-34). Oxford:
Oxford University Press.
Lee, J. (2010). Usability of the primary measures of music audiation (PMMA) with 5-year-old
Korean children (Doctoral dissertation). Retrieved from ProQuest Dissertations and
Theses database. (UMI No. 3407979)
Levinowitz, L. M. (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.

	
  

155	
  

Levinowitz, L. M., & Guilmartin, K. K. (2000). Music together. In Spotlight on early childhood
music education (pp. 15-16). Reston, VA: MENC, The National Association for Music
Education.
Malbrán, S. (2002). Tapping in time: A longitudinal study at the ages of three to five years.
Bulletin of the Council for Research in Music Education, 153/154, 71-77.
McAfee, O., & Leong, D. (2002). Assessing and guiding young children's development and
learning. Boston, MA: Allyn and Bacon.
McCusker, J. (2001). Emerging musical literacy: Investigating young children's music cognition
and musical problem-solving through invented notations (Doctoral dissertation).
Retrieved from ProQuest Dissertations and Theses database. (UMI No. 3004935)
McDonald, D. T., & Simons, G. M. (1988). Musical growth and development: Birth through six.
New York, NY: Schirmer Books.
Metz, E. (1989). Movement as a musical response among preschool children. Journal of
Research in Music Education, 37(1), 48-60.
Michigan Education Association [MEA]. (2011). Teacher evaluation and pay for performance.
East Lansing, MI: Author. Retrieved from http://www.mea.org/
Miller, L. B. (1986). A description of children's musical behaviors: Naturalistic. Bulletin of the
Council for Research in Music Education, 87, 1-16.
Miller, E., & Almon, J. (2009). Crisis in the kindergarten: Why children need to play in school.
College Park, MD: Alliance for Childhood.
Moog, H. (1976). The musical experience of the pre-school child. (C. Clarke, Trans.). London:
Schott Music.
Moorhead, G. E., & Pond, D. (1978 [1941-1951]). Music of Young Children. Santa Barbara, CA:
Pillsbury Foundation for the Advancement of Music Education.
Nash, G. C. (1974). Creative approaches to child development with music, language, and
movement. Port Washington, NY: Alfred Pub. Co.
National Association for the Education of Young Children & National Association of Early
Childhood Specialists in State Departments of Education [NAEYC & NAECS/SDE]
(2003). A joint position statement on early childhood curriculum, assessment, and
program evaluation: Building an effective, accountable system in programs for children
birth through age 8. Retrieved from http://www.naeyc.org/positionstatements/cape
National Association for Music Education [NAfME] (2012a). Performance standards for music:
Assessment strategies for music. Reston, VA: National Association for Music Education.

	
  

156	
  

Retrieved from http://musiced.nafme.org/resources/performance-standards-formusic/assessment-strategies-for-music/
National Association for Music Education [NAfME] (2012b). The school music program: A new
vision. Reston, VA: National Association for Music Education. Retrieved from
http://musiced.nafme.org/resources/the-school-music-program-a-new-vision/
National Council on Teacher Quality [NCTQ] (2012). State of the states 2012: Teacher
effectiveness policies. Washington, DC: Author. Retrieved from
http://www.nctq.org/dmsView/State_of_the_States_2012_Teacher_Effectiveness_Policie
s_NCTQ_Report
National Research Council [NRC]. (2001). Eager to learn: Educating our preschoolers. In B. T.
Bowman., M. S. Donovan., & M. S. Burns (Eds.). Washington, DC: National Academy
Press.
Neisworth, J. T., & Bagnato, S. J. (2004). The mismeasure of young children: The authentic
assessment alternative. Infants and Young Children, 17(3), 198-212.
Piaget, J. (1950). The psychology of intelligence. New York: Harcourt, Brace.
Puckett, M. B., & Black, J. K. (2000). Authentic assessment of the young child: Celebrating
development and learning (2nd ed.). Upper Saddle River, NJ: Merrill.
Rainbow, E. (1981). A final report on a three-year investigation of the rhythmic abilities of
preschool aged children. Bulletin of the Council for Research in Music Education, 66/67,
69-73.
Rainbow, E. L., & Owen, D. (1979). A progress report on a three year investigation of the
rhythmic ability of pre-school aged children. Bulletin of the Council for Research in
Music Education, 59, 84-86.
Rauscher, F. (2000). Is assessment in music appropriate in the early childhood years? In
Spotlight on early childhood music education (pp. 49-51). Reston, VA: MENC, The
National Association for Music Education.
Ravitch, D. (2010). The death and life of the great American school system: How testing and
choice are undermining education. New York: Basic Books.
Reynolds, A. M. (1995). An investigation of the movement responses performed by children 18
months to three years of age and their caregivers to rhythm chants in duple and triple
meters (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses
database. (UMI No. 9527531)
Runfola, M., & Rutkowski, J. (Eds.). (2010). Tips: The child voice. Lanham: Rowman &
Littlefield Education.

	
  

157	
  

Rutkowski, J. (1990). The measurement and evaluation of children's singing voice development.
The Quarterly, 1(1-2), 81-95. [Reprinted with permission in Visions of Research in Music
Education, 16(1), 2010]. Retrieved from http://www-usr.rider.edu/~vrme/
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(4), 353-368.
Rutkowski, J. (2010). The singing voice development measure: Assessing children's use of
singing voice in the elementary general music class. In T. S. Brophy (Ed.), The practice
of assessment in music education: Frameworks, models, and designs. Proceedings of the
2009 Florida Symposium on Assessment in Music Education. (pp. 273-280). Chicago, IL:
GIA Publications.
Rutkowski, J., & Miller, M. S. (2003). The effectiveness of frequency of instruction and
individual/small group singing activities on first graders' use of singing voice and
developmental music aptitude. Contributions to Music Education, 30(1), 23-38.
Schleuter, S. L., & Schleuter, L. J. (1985). The relationship of grade level and sex differences to
certain rhythmic responses of primary grade children. Journal of Research in Music
Education, 33(1), 23-29.
Seefeldt, C. (1998). Assessing young children. In C. Seefeldt & A. Galper (Eds.), Continuing
issues in early childhood education (pp. 314-338). Upper Saddle River, NJ:
Merrill/Prentice Hall.
Shelley, S. J. (1981). Investigating the musical capabilities of young children. Bulletin of the
Council for Research in Music Education, 68, 26-34.
Shepard, L., Kagan, S. L., & Wurtz, E. (Eds.). (1998). Principles and recommendations
for early childhood assessments. Washington, DC: National Education Goals Panel.
Simons, G. (1986). Early childhood musical development: A survey of selected research.
Bulletin of the Council for Research in Music Education, 86, 36-52.
Sims, W. L. (1985). Young children's creative movement to music: Categories of movement,
rhythmic characteristics, and reactions to changes. Contributions to Music Education, 12,
42-50.
Sloboda, J. A. (1985). The musical mind: The cognitive psychology of music. Oxford: Clarendon
Press.
Smale, M. J. (1987). An investigation of pitch accuracy of four-and five-year-old singers
(Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses database.
(Order No. 8723851)

	
  

158	
  

Strauss, A, L., & Corbin, J. M. (1990). Basics of qualitative research: Grounded theory
procedures and techniques. Newbury Park, CA: Sage Publications.
Sullivan, K. N. (2006). The responses of preschool children to early childhood music instruction
containing texted and non-texted songs and chants (Master's thesis). Retrieved from
ProQuest Dissertations and Theses database. (UMI No.1434373)
Sundin, B. (1997). Musical creativity in childhood - A research project in retrospect. Research
Studies in Music Education, 9, 48-57. doi: 10.1177/1321103X9700900106
Taggart, C. C. (1989). The measurement and evaluation of music aptitudes and achievement. In
D. L. Walters & C. C. Taggart (Eds.), Readings in music learning theory (pp. 45-53).
Chicago, IL: GIA Publications.
Taggart, C. C. (1994). A validity study of "Audie": A test of music aptitude for 3- and 4-year-old
children. Bulletin of the Council for Research in Music Education, 121, 42-54.
Taggart, C. C. (2000). Developing musicianship through musical play. In Spotlight on early
childhood music education (pp. 23-26). Reston, VA: MENC, The National Association
for Music Education.
Taggart, C. C. (2005). Meeting the musical needs of all students in elementary general music. In
M. Runfola & C. C. Taggart (Eds.), The development and practical application of music
learning theory (pp. 127-142). Chicago, IL: GIA Publications.
Taggart, C. C., Alvarez, J., & Schubert, K. (2011). The role of early childhood music class
participation in the development of four children with speech and language delay. In S. L.
Burton & C. C. Taggart (Eds.), Learning from young children: Research in early
childhood music (pp. 245-258). Lanham, MD: Rowman & Littlefield Education.
Taggart, C. C., Reynolds, A. M., Valerio, W. H., Bailey, J. M., Lange, D., & Gordon, E. E.
(2009). Jump right in: Kindergarten. Chicago, IL: GIA Publications.
Valerio, W. H., Reynolds, A. M., Bolton, B. M., Taggart, C. C., & Gordon, E. E. (1998). Music
Play: The early childhood music curriculum guide for parents, teachers, and caregivers.
Chicago, IL: GIA Publications.
Veldhuis, H. A. (1992). Singing of preschool children in two contexts (Doctoral dissertation).
Retrieved from ProQuest Dissertations and Theses database. (Order No. 9300151)
Welch, G. F. (1994). The assessment of singing. Psychology of Music, 22, 3-19. doi:
10.1177/0305735694221001
Welch, G. F. (1998). Early childhood musical development. Research Studies in Music
Education, 11, 27-41. doi: 10.1177/1321103X9801100104

	
  

159	
  

Welch, G. F., Sergeant, D. C., & White, P. J. (1996). The singing competencies of
five-year-old developing singers. Bulletin of the Council for Research in Music
Education, 127, 155-162.
Whitcomb, R. (2010). Rhythmic characteristics of improvisational drumming among preschool
children. Research and Issues in Music Education, 8(1). Retrieved from
http://www.stthomas.edu/rimeonline
Woods, D. G. (1989). The music experience laboratory: A model. In B. Andress (Ed.),
Promising practices: Prekindergarten music education (pp. 1-11). Reston, VA: Music
Educators National Conference.
Wright, S. K. (1994). Assessment in the arts: Is it appropriate in the early childhood years?
Studies in Art Education, 36(1), 28-43.
Young, S. (2002). Young children's spontaneous vocalizations in free-play: Observations of twoto three-year-olds in a day-care setting. Bulletin of the Council for Research in Music
Education, 152, 43-53.
Young, S. (2003). The interpersonal dimension: A potential source of musical creativity for
young children? [special issue]. Musicae Scientiae, 175-191.
Zemke, L. (2000). Assessing musical growth in early childhood. In Spotlight on early
childhood music education (pp. 52-53). Reston, VA: MENC, The National Association
for Music Education.

	
  

160