I A.‘

 
 

WWW}

  

NW

i

   

NW

|

M

   

WW

   

W!

x
‘

    
 

3,3032 W

     

TH

'0)

THESIS
90-03

This is to certify that the
thesis entitled
AN INVESTIGATION OF THE EFFECTS OF
AURAL MODELS ON THE DEVELOPMENT OF BEGINNING
INSTRUMENTALIST'S TONE PRODUCTION AND STYLISTIC
UNDERSTANDING
presented by

Peter Owen Conniff

has been accepted towards fulfillment
of the requirements for

M degree in Wtion

6 Major profess

Date <3 L13 1%?

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY
Michigan State
University

 

 

 

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

 

DATE DUE I DATE DUE DATE DUE

 

:I 7 31m '1 2 2004]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65«p.15

An Investigation of the Effects of Aural Models
on the Development of Beginning ‘Instrumentalist's
Tone Production and Stylistic Understanding
By
Peter Owen Conniff

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF MUSIC EDUCATION
School of Music
1999

ABSTRACT

An Investigation of the Effects of Aural Models
on the Development of Beginning Instrumentalist’s
Tone Production and Stylistic Understanding
By

Peter Owen Conniff

The purpose of this study was to investigate the effects of aural models
on the development of elementary students’ performance skills. There are
two specific problems. First, will students develop a more characteristic tone
on their instrument if, during home practice, they listen to an aural model of
a professional playing the same instrument that they are studying, and
second, will students develop a better sense of phrasing (style) if, during
home practice, they listen to an aural model of a professional playing the
same instrument that they are studying?

The study was a treatment] control group design. The fifth-grade
students (N=56) were randomly assigned to the treatment or control group.
The treatment group received a home listening tape of professional
instrumentalists playing songs on the same instruments that they were
studying. They were to listen to the tape as part of their practice routine. At
the conclusion of 12 weeks, the tone quality and phrasing ability of students
in both groups were evaluated using rating scales designed for each
dimension.

The researcher found no significant differences between the treatment
and control groups’ scores. However, this finding does not agree with the
body of research on modeling.

ACKNOWLEDGMENTS

My appreciation is expressed to Dr. Cynthia Taggart, committee
chairperson, for her guidance, support, and encouragement throughout my
degree program and during the process of completing this thesis.

I would like to express my thanks to my committee members Dr. John
Kratus, and Dr. Gordon Sly for their input toward the preparation of this
project and writing of this document. Many thanks to Katherine Berry for her

counsel, help and participation in implementing this study.

iii

TABLE OF CONTENTS

CHAPTER PAGE
I. INTRODUCTION ........................................................................ 1
Modeling ........................................................................... 1
Modeling and Music Preference .................................. 1
Modeling and Student Performance ........................... 2
Summary .......................................................................... 6
Purpose and Problems .................................................... 7
11. REVIEW OF RELATED LITERATURE .................................. 8
Introduction ..................................................................... 8

Use of modeling with Advanced Instrumentalists . 8
Use of Modeling With Younger Instrumentalists 12

Summary .......................................................................... 17

III. METHOD ....................................................................................... 21
Subjects and Setting ........................................................ 21

Procedure .......................................................................... 21

Rating Scales ..................................................................... 25

Analysis of the Data ........................................................ 27

IV. RESULTS ....................................................................................... 29
Inter judge Reliability Results ...................................... 29

Means and Standard Deviations .................................. 29

Results for Tone Quality ................................................ 30

Results for Phrasing ........................................................ 30
Interpretation of results ................................................. 31
V. SUMMARY, CONCLUSIONS, RECOMMENDATIONS, AND
IMPLICATIONS ............................................................................ 35
Summary .......................................................................... 35

Conclusions ...................................................................... 36
Recommendations .......................................................... 37

Implications ...................................................................... 39
APPENDICES ....................................................................................................... 40
Appendix A ............................................................................................. 41
Appendix B .............................................................................................. 42

Appendix C, D, and E (Judging materials) ......................................... 44-6

LIST OF REFERENCES ....................................................................................... 47

iv

LIST OF TABLES

TABLE PAGE
4.1 Means and Standard Deviations .................................................... 30
4.2 Results of T-test on Tone Quality Mean Scores .......................... 30
4.3 Results of T-test on Phrasing Mean Scores .................................. 31

LIST OF FIGURES

FIGURES PAGE
3.1 Tone Quality Rating Scale ............................................................... 26
3.2 Point Assignments for first phrase of post-test selection .......... 27

vi

CHAPTER I
INTRODUCTION

Modeling

Teachers have traditionally used many means of conveying
information to their students. Among these are lecturing, explanation, and
demonstration or modeling. Of these three, modeling has been shown
through research to be the most effective (Anderson, 1981; Brand, 1977;
Davidson, 1989; Dickey, 1989; Fortney, 1992; Gonzo and Froseth, 1976;
McCarthy, 1974; Moore, 1976; Poupolo, 1970; Rosenthal, 1984; Rosenthal et. a1.
1988; Sang, 1987; Zucher, 1975). A model, defined by Webster's New World
Dictionary (1984), is ”a person or thing considered as a standard of excellence
to be imitated." Modeling, then, is the act of presenting that standard of
excellence through demonstration. Leonard and House (1941 pp. 236-37) stress
that students must have an [aural] concept of their performance goal to guide
their efforts during individual practice. In other words, students need to have
a model.

In relation to instrumental music education, the aural concept students
need is the sound of the particular instrument he or she is studying. In this
case, the standard of excellence is considered to be a musician that can play an
instrument with excellent quality sound and style.

Music Preference and Mocglinjg

Baker (1980) and Peery and Peery (1986) have looked at how the models
that young children have during music instruction can affect their preference
for different musical styles. Peery 8t Peery (1986) found that preschool-aged
children exposed to classical music consistently over time showed an

increased preference for that type of music over popular music. They also

found that preference for classical music from pre- to post-testing atrophied
with students who were not exposed consistently.

Baker (1980) worked with third and fourth grade students in the
classroom setting to see if their preference for lullabies and capstan chanteys
would be affected by appropriate and inappropriate models. He found that
those students who were taught the songs with an appropriate style (lullabies
sung soft and slow and capstan chanteys sung fast and loud) preferred to
listen and perform the songs with that style, while the students who were
taught with an inappropriate song style (lullabies sung fast and loud and
capstan chanteys sung soft and slow) preferred to listen and perform the songs
with that style. Peery 6t Peery and Baker’ 3 research support the idea that
children will accept what is taught to them as correct and appropriate. If that
is true, and music teachers want children to perform with an appropriate
sound on their instruments, a model that demonstrates a correct and
appropriate sound should be used to teach young musicians. If an
inappropriate model is used, music educators may be teaching students to
prefer that inappropriate sound and, later, not know why they are not playing
with a better sound or sense of style.

Modeling and Student Performance

If teacher modeling can affect student musical preferences, could they
also affect student performance? Several researchers have investigated
modeling to determine if it is a superior method to the more conventional
lecture and drill style teaching with college aged students (Brand, 1977 ;
Davidson, 1989; Rosenthal, 1984; Rosenthal et. a1. 1988), middle and high
school aged students (Dickey, 1989; McCarthy, 1974), and elementary aged
students (Anderson, 1981 ; Fortney, 1992; Gonzo and Froseth, 1976; Moore,
1976; Poupolo, 1970; Sang, 1987; Zucher, 1975).

With college aged students, achievement success was seen as a direct
result of teacher modeling. In a descriptive study by Davidson (1989), a master
teacher in China was observed teaching a lesson. Davidson notes that after
the master teacher modeled musical gestures, the student played more
musically. Davidson stresses that the important factor in the student’ 5 success
was the master teacher’ 3 ability to present a high quality model. Brand (1977)
examined college professor’ 5 teaching methods used in college level
classroom-management classes. He found that when teacher management
behaviors and techniques were modeled instead of taught through verbal
lecture, students showed more effective management skills in actual teaching
situations.

Rosenthal (1984) was concerned with models presented aurally .
through an audio cassette-tape recording. She designed an experimental study
in which college students were presented with a performance task They were
given a short amount of time to learn an unfamiliar piece of music.
Available to the student for guidance was a model communicated with
varying types of lecture/ performance combinations in the form of an audio
cassette-tape recording. Each group in the study had a different combination
of lecture and performing on their cassette tape recording. Rosenthal found
the group that performed the best had the model only tape. This suggests that
even though a teacher is not present, an aural model alone can facilitate
student performance. Rosenthal et. a1. (1988) continued research with the
effectiveness of using an aural model during practice by comparing it to silent
practice, free practice, and singing. Of the four ways of practicing the
unfamiliar material, the students in the modeling group performed better
than the students in the other three groups. This finding supports the above

study that aural models can help student performance and suggests further
that it is more effecu've than other forms of practicing.

On the middle school level McCarthy (1974) observed traditional
instruction methods used by many band directors and found that, ”The
instructional procedures in [instrumental] classes typically involve brief
periods of lecture/ explanation/ demonstration by the instructor, followed by
extensive simultaneous drill for all students on the identical exercise.” Most
of the above procedures involve the students in an activity that does not give
them the opportunity to hear what they should sound like. Even though they
may have heard the model at the beginning of the class, there needs to be
continual presentation of the model to guide student efforts. Dickey’ s (1989)
findings support the work of McCarthy. He found improved melodic and
rhythmic discrimination skills with eighth grade instrumental students after
a treatment that involved teacher modeling and student imitation.

On the elementary level, Anderson (1981) looked at the use of
modeling without verbal guidance in home practice sessions of sixth grade
beginning band students. Dependant variables in this study were pitch
reading, rhythm reading, tempo accuracy, and intonation accuracy. He found
no significant differences between treatment and control groups for any of the
dependant variables. His results did not support the research of the above
mentioned studies, but did continue to develop the methodology for
conducting such a study. Fortney (1992) also researched the affects of
modeling without verbal instruction in a practice situation with different
results. A design closer to that of Rosenthal (1988) was used. Each student was
given a short amount of time to practice an unfamiliar piece of music and
then asked to perform it for the researcher. He found that when a student
listens to the selection to be played, the resulting performance is better than if

a student is allowed free practice, is asked to silently analyze the music, or is
asked to sight read a selection without practice before the performance of the
piece.

Fortney’s (1992) findings support the earlier work of Poupolo (1970).
Poupolo gave students in his experimental group a cassette tape that directed
them through a practice session. The session included some verbal
instruction followed by performances of the exercises to be learned. The
student was to listen to the model and make an attempt to play the exercise.
Students in the control group conducted self-guided practice sessions.
Poupolo found that the students with the programmed practice sessions that
included many aural model presentations scored better than students who
conducted self guided practice. Zucher (1975) continued the work of Poupolo
with programed practice sessions. He added sections in the programed tape
for students to play along with the models to develop the students error
detection during performance. As with Poupolo’s study, Zucher also found
that higher performance scores were attained by students using the A
programed practice tapes.

The studies discussed above are concerned with models presented by a
video- or audio-tape recording. During band rehearsal or private lesson time,
an instructor often will provide a live model for the students. This requires
that the teacher has some performance skill on each instrument he/ she is
attempting to model. Looking more closely at what skills the teachers possess,
Sang (1987) rated first year teachers on their ability to play secondary
instruments and compared these ratings to the performance scores of their
students after one year of instruction. There was a significant correlation

between the teacher ratings and the student performance scores, suggesting

that the higher quality a model teachers are able to present, the higher
performance skills their students are able to achieve.
Summag

Every student in music classes possesses a music aptitude that may or
may not be realized in that student’s current level of achievement. This puts
more of the responsibility on the educator to develop better ways to attend to
the learning needs of students. It is not acceptable for a teacher to blame a
student’ 3 lack of musical ”talent” for poor achievement. Research has shown
that the experience and ability of the teacher affect the achievement of a
student. Also, a teacher’ 5 ability to present a quality model in the classroom
may be the key to a student’ 5 success.

One problem instrumental music teachers face is the number of
models they need to be able to present. Many teachers are not prepared to
model effectively on every instrument that students are learning. One
possible solution to this dilemma might be to have students routinely
listening to excellent models via tape recordings of professional players. This
might provide the good models needed to develop a concept of quality
playing that, when combined with the usual and necessary verbal lecture and
explanation during classroom experiences, could nurture excellent
instrumentalists. Current research should be concerned with finding ways to
present these models in the practical educational setting. If educators do not
embrace this challenge, their student’s preference for musically appropriate
playing might be compromised, and students might never realize the

outstanding musical achievement their teachers aspire to facilitate.

Purpose and Problems

The purpose of this study was to investigate the effects of aural models
on the development of elementary students’ performance skills. The
problems of this study are as follows:

1) Will students develop a more characteristic tone on their
instrument if, during home practice, they listen to an aural model of a
professional playing the same instrument that they are studying?

2) Will students develop a better sense of phrasing (style) if, during
home practice, they listen to an aural model of a professional playing the

same instrument that they are studying?

CHAPTER II
RELATED LITERATURE
Intrcfiuction

Presenting tape recorded models in the educational setting is not a new
idea. Researchers have investigated the effects of tape-recorded models on
performance skills such as rhythm and pitch accuracy, phrasing and
articulation, dynamics and tempo (Rosenthal, 1984; Rosenthal et. al., 1988;
Poupolo, 1970; Zucher, 1975; Fortney, 1992) This chapter includes discussion
of these related studies.

Use of Modeling vLiLh Advanced Instrumentalists

Rosenthal (1984) was concerned with instruction that involved both
verbal and musical models. The purpose of this study was to compare the
success of musical and verbal models, used alone and in combination during
practice sessions, on musicians’ performance. Verbal models were termed
”guides” throughout the study.

The subjects in this study consisted of 44 graduate and upper level
undergraduate students studying a woodwind or brass instrument. The
students were randomly assigned to one of four treatment groups. Each
group's treatment consisted of a different combination of verbal guides and
musical models; the groups were Guided Model, Model Only, Guide Only,
and Practice Only, in which no verbal guide or musical model was presented.

The subjects were asked to tape record themselves performing an
unfamiliar etude after a 10 minute practice session. During the practice
session, the Practice Only group, acting as the control, was instructed to
practice simply the etude with no outside guidance or no model for the full 10
minutes. The Guide Only group received a cassette tape with a verbal
explanation of the etude, the Model Only group received a cassette tape with

the etude performed three times by a professional violinist, and the Guided
Model received a cassette tape with combined verbal explanation and
performed excerpts of the etude. After listening to the cassette tape, the
students had three minutes to practice before recording the etude. Two
independent judges rated each subject’s performances by comparing the
recordings to a score of the etude. The judges evaluated pitches, rhythms,
articulation/ phrasing, dynamics, and tempo.

The results of the study showed significant differences for all
dependent variables except articulation] phrasing. The highest scores were
achieved by the Model Only group, followed by the Guided Model group, and
the lowest scores were achieved by the Practice Only and Guide Only groups.

Rosenthal concluded that different modeling conditions can affect
subjects’ performance. Rosenthal noted that, although the Guided Model was
designed to be most effective, as it attempted to highlight the critical areas of
the piece, the guide may have actually hindered the student’ 3 ability to
perform accurately the etude. The study shows that pure modeling, without
any added verbal explanation, is most effective in music instruction over a
short period of time and that verbal explanation alone is no better than
unguided practice. Rosenthal suggests that if verbal explanation must be used,
it should be used in conjunction with modeling.

Rosenthal clearly showed the short term benefits to using an unguided
model of the material being studied by the student. Three questions arise for
future research from the results of this study. First, this study used advanced
college level musicians as subjects. Would the same results be gained by using
a population of elementary-aged beginning band students? Second, Rosenthal
addressed pitches, rhythms, articulation/ phrasing, dynamics, and tempo but
not tone quality. Would the use of recorded models have an effect on

students’ tone quality as well? Third, Rosenthal’s models were meant to
present exactly what the student was required to perform at the end of the
treatment period. Would a student, who listened to a model presenting the
proper tone quality of the student’s particular instrument and phrasing of the
same type, be able to generalize what they have learned aurally to new music?

Rosenthal, Evans, Greenwalt, and Wilson (1988) conducted a study
comparing different practice conditions including modeling. The purpose of
this study was to determine the relative effectiveness of silent practice,
modeling, singing, and free practice in helping music students to master a
musical selection quickly. The subjects for this study were sixty graduate and
upper-level undergraduate woodwind and brass majors attending a college of
music. Each subject was randomly assigned to one of five treatment groups
and worked with the researchers individually throughout the study.

When each subject entered the research room, he or she was presented
with an etude and given appropriate instructions on how to practice,
depending on his/ her assignment to a treatment group. Members of the
modeling group were instructed to listen to a recording of the etude while
they watched the score. The singing group members were instructed to learn
the etude by singing it through in a full voice using a keyboard to find pitches
as necessary. The silent analysis group members were instructed to visually
study the music for 3 minutes. The free practice group members were
instructed to continually play through the etude for 3 minutes. Finally,
members of the control group were given an unrelated etude to practice for
the 3 minutes. At the end of the practice session the subjects were asked to
tape record themselves sight reading the experimental etude to the best of
their ability. Each recorded performance was evaluated in terms of note,

rhythm, and articulation accuracy on a beat by beat basis, in terms of phrasing

10

or dynamics on a right or wrong basis, and in terms of tempo on the basis of
the number of seconds student performance time deviated from the model’s
performance time.

The results indicated significant differences in performance of rhythm,
phrasing, dynamics, and tempo and indicated non—significant results in
performance of notes and articulation. The subjects in the modeling group
attained the highest scores in all areas except rhythm. Rosenthal et. a1.
concluded the most important finding to be that, for quick learning, simply
hearing the model without practice was just as effective as practicing without
hearing the model.

Rosenthal et. al. have shown modeling to be a effective way to convey
musical information to college-aged student over a short period of time.
Three questions arise after the review of this study. First, would a practice
session that includes listening to a model in addition to the opportunity for
practice increase the quality of the recorded performance? Second, would a
practice condition that includes listening to a recorded model over several
practice sessions benefit the student to a greater degree than a one-time, three
minute practice session? And third, would the same results have been seen if
the subjects had been elementary-aged students?

There is research that has shown modeling to be beneficial for
elementary-aged students. Studies by Poupolo, 1970; Zucher, 1972; and
Fortney, 1992 that focus on modeling and younger musicians are discussed

below.

11

Use of Modelingwith Younger Instrumentalists

Poupolo (1970) was concerned about the lack of guidance students
receive during individual practice time. This study investigated how
programmed instruction using cassette tapes would increase the efficiency of
the individual practice session. The two objectives most relevant to the
present study were 1) to develop self-instructional practice material on
recorded tape from assigned lesson material and 2) to test the feasibility of
structured practice with this new material to aid learning of concepts and
developing skills for performance on a musical instrument. Poupolo set out
to invalidate many null hypotheses. Most relevant to the present study was
the null hypothesis that structured practice with recorded tapes of
programmed material would not produce a significant difference in
performance achievement.

The subjects for this study were 52 fifth-grade beginning trumpet and
cornet students from six elementary schools. The sample was all boys taught
by the same teacher; factors of gender and instructor influence were not
involved in the analysis. Subjects were pre-tested in terms of music
achievement, social status, and IQ, then randomly assigned to either the
experimental or control group. The effects of social status and IQ on the
results of this study were included in null hypotheses but are not relevant to
the present study. Therefore, they will not be discussed here. For the next ten
weeks, subjects participated in the band program, but each group had different
practice instructions. Subjects in the experimental group practiced each day
for the next ten weeks with practice tapes that guided each practice session.
The practice tapes included a model performance of the weekly material,
opportunities for student response, reinforcement listening, and verbal

instructions and reminders. Piano accompaniments were used during model

12

performance, student response, and reinforcement listening sections of the
practice tape. The control group practiced the same material as the
experimental group but without the practice tapes. Instead, their practice
session was self-directed and lasted as long as the experimental group
sessions. At the conclusion of the ten weeks, both groups were post-tested
using the Watkins-Famum Performance Scale. This is a series of sight
reading exercises that get progressively more difficulty. One point is awarded
for each measure of the exercise played correctly. The measure is considered
correct if all the following factors are performed without error: the factors are
pitch, tempo, length of note, expression, slurs, rests, pauses, and repeats. Any
error negates the point for that measure.

The results indicated that programmed practice produced superior
performance to non-programmed practice. Also, subjects in the
programmed-practice group who pre-tested with above-average music
achievement showed significantly better performance achievement than
those who pre-tested with below-average music achievement. Subjects in the
non-programmed practice group showed no significant difference in
performance achievement between subjects who pre-tested with above- and
below-average music achievement. Poupolo (1970) concluded that
programmed practice can improve the performance of all elementary
students, regardless of their current music achievement. He also found this
type of guided practice, with every step of the session cued verbally and the
music to be learned presented with an aural model, to be a more efficient way
to improve the performance of young musicians.

After review of this study, two issues are noted by the researcher. First,
by using elementary-aged students who are just beginning their instrumental
instruction, Puopolo increases the meaning of his results for public school

13

music educators. However, Poupolo’s sample included only boys studying
brass instruments which does not reflect a normal and diverse public school
population. Would the same results be found, then, using a diverse -
population of students including girls and boys studying woodwind and brass
instruments? Second, cassette tapes are used to provide performance models
for individual practice. They also present the material being studied by the
students with verbal instructions to guide their practice. Could practice tapes
with no verbal directions, as used by Rosenthal et. al. (1988), and material not
being studied by the student improve the performance of elementary-aged
students? In other words, can students generalize the musical information on
the tapes to the selections they are working on in band.

Zucher (197 5) continued the work of Poupolo (1970) by looking at
programmed instruction in relation to students’ abilities to detect errors and
adjust during simultaneous feedback instead of during confirmation
feedback. Simultaneous feedback occurs when the student plays the tune
along with a recorded model and is required to aurally check his or her
performance with the model. In confirmation feedback, the student listens to
the model after he or she has performed the tune to determine if there were
any mistakes. The purpose of this study was to compare model-supported
practice (similar to Poupolo’s programmed practice sessions) and traditional
self-directed practice in terms of gross pitch discrimination (exceeding +/ - 100
cents), tempo stability, pitch matching, fingering and slide position errors,
rhythm errors, and time spent in practice.

The subjects for this study were 43 beginning fourth- fifth— and sixth-
grade trumpet, horn, trombone and baritone students enrolled in their public
school band programs. Each subject was randomly assigned to an
experimental or control group during the first week of the study. Every

14

following week, the groups rotated treatments. This rotation allowed both
weekly and cumulative measurements to be included in the results. The
study ran for 6 weeks.

Students had a 15 minute lesson each week to test the mastery of the
previous week’s material, make corrections on that material, collect the tapes
and books from the previous week, assign new material, and give out practice
tapes and books for new material. Students were held accountable for their
practice time by two methods. First, fifteen students were randomly selected
to tape record a few of their practice sessions, and second, all students noted
their practice times on a practice chart. The accuracy of the chart was verified
by a parent signature. A reward system was used to promote honesty in
practice time reporting and attendance at the scheduled lessons. One token
was given for attending, another for bringing everything they needed, and
another for having a complete and verified practice record. Three tokens
resulted in their being given a bag of candy at the end of the lesson.

The results of the study showed that model-supportive practice is more
effective than traditional practice in relation to gross-pitch discrimination,
pitch matching, rhythmic accuracy, and time spent in practice. Through the
analysis of the data, Zucher concluded that achievement was not affected by
the amount of time spent practicing. He suspected that many students
increased their practice time because the length of the tape was longer than
their normal practice session. Still, it was the increased quality of practice time
due to the practice tapes that influenced the improvements shown in the
results rather than the increased quantity.

These findings support the present study because they show that
presenting a quality model in an individual practice session is an effective

way to improve a student’ 3 performance achievement. Three issues are noted

15

here by the researcher. Zucher showed that students have the capability to
discriminate and adjust for pitch differences in their playing as compared to
the tape recorded model. Could students exhibit the same aural
discrimination with respect to tone quality and phrasing? Second, the use of
only brass players for the population is a limiting factor in this study’ 3 results.
Although it is broader than that of Poupolo, who used only trumpet and
cornet students, future studies could utilize a population more representative
of public school band programs. And third, the study lasted 6 weeks. Would
the benefits increase if the treatments were used for a longer period of time?

Fortney (1992) was also concerned with modeling as a teaching
technique for elementary instrumental students. The purpose of his study
was to determine the relative effectiveness of modeling and silent analysis on
the performance ability of advanced elementary school instrumentalists.

The subjects in his study were 40 second-year, elementary-school
instrumentalists enrolled in the band program. They were randomly assigned
to one of four groups: Modeling, silent analysis, free practice, and control. All
students were asked to sight read an unfamiliar etude as a pretest measure.
The researcher continued with the following procedure.

The subjects in the modeling group were instructed to listen to a tape
recording of the etude, practice for two minutes, and then record themselves
performing the etude. The subjects in the silent analysis group were
instructed to rehearse the etude mentally for two minutes while looking at
the score and then record their performance. The subjects in the free practice
group were instructed to practice the etude on their instruments for two
minutes and then record their performance. The subjects in the control group
practiced an unrelated etude for two minutes and then were asked to sight
read the etude. The recordings were scored by giving one point for each

16

accurately played measure. The dependent variables were pitch, rhythm, and
articulation. Sound quality, interpretation, and dynamics were not examined.
If any of the variables in a given measure were played incorrectly no point
was awarded for that measure.

The results were based on the gain scores from pre- to post-test. The
modeling groups showed the greatest gains, followed, in descending order, by
the silent analysis, free practice, and control groups. The modeling group
scores were the only scores that showed a significant difference from the
control group scores. Therefore, he concludes that modeling was the most
effective practice method. He suggested that modeling should be used by band
instructors to help make their students’ practice sessions more meaningful
and productive.

Fortney’s study is similar to the Rosenthal (1984) and Rosenthal et. a1.
(1988) studies in that both use tape recorded models during a one time
practice session. He showed that the results shown in college-aged students
can be seen in elementary-aged students. Would these results be the same if a
practice condition similar to Fortney’s was used by students over a longer
period of time? Fortney was concerned with pitch, rhythm, and articulation,
and did not focus on tone quality or phrasing. Would the same practice
conditions produce the same results with respect to tone quality and phrasing
and, if positive, would those results be magnified if the practice conditions
were used over a longer period of time?

Summary

These studies show that modeling is an effective method of teaching
instrumental music to students. Rosenthal (1984) was concerned with the
effects of models and verbal guides, alone and in combination, on student

performance. Performances were scored in relation to accuracy of notes,

17

rhythms, articulation, phrasing, dynamics and tempo. Using the same
dependent variables, Rosenthal et. al. (1988) studied how a practice condition
using modeling compared to the practice conditions of silent analysis, free
practice and singing. Poupolo (1970) was concerned with programmed practice
during home practice sessions. He developed a practice tape that included
verbal instruction guiding each step of the practice session interspersed with
models of each exercise. Students were evaluated on their ability to sight read
with accurate pitch, rhythm, expression, and to follow music symbols such as
repeats and slurs. Zucher (1975) was also concerned with guided home
practice tapes on which the entire session was verbally directed with aural
models of the exercises to be learned. He evaluated his students rhythm
accuracy and, unlike the other researchers mentioned here, on gross pitch
discrimination, pitch matching, and time spent in practice. Fortney (1992)
returns to similar questions posed by Rosenthal et. al. (1988) and compared
the practice conditions of modeling, silent analysis, free practice, with a
control group whose members sight read their performances. None of these
researchers looked at the effects of modeling on tone quality development.
In all five of these studies, slightly different populations were used.
Rosenthal (1984) used 44 subjects. Male and female college students learning
woodwind and brass instruments were included in the study. They were
studying at the graduate and advanced-undergraduate level. Rosenthal et. al.
(1988) used the same type of population except that the number included in
the study was increased to 60. Poupolo (1970) included fifth—grade trumpet
and cornet players just beginning their instrumental instruction. All of the 52
subjects were boys. Zucher (1975), like Poupolo (1970), used only brass
students but included both girls and boys. He also expanded the grade-level
range of the students to fifth- and sixth-graders. Fortney (1992) had a reduced

18

number of subjects having only 40 and, unlike Poupolo (1970) and Zucher
(197 5), he included both woodwind and brass instrumentalists. Departing
from both Poupolo(1971) and Zucher (1975), Fortney (1992) used students in
their second year of study. None of these researchers used a first year
population that included both boys and girls who were learning both
woodwind and brass instruments. This population seems most similar to one
a normal public-school band program might have, lending to the
generalizability of results.

Each of these researchers designed a study using varying practice
conditions. Rosenthal (1984) had her subjects practice for 10 minutes using
the prescribed practice treatment before the performance was recorded. The
subjects were scored on the performance of the music they had practiced.
Rosenthal et. al. (1988) again used a one time practice session but reduced the
time allotted to 3 minutes before a performance was recorded. The students
were scored on the performed of the music that they had practiced. Fortney
(1992) also used a short, one-time practice session before the performance was
recorded. He further reduced the length of the practice session to 2 minutes.
Again the students were scored on the performance of the music that they
had practiced. Zucher (1975) used a much longer treatment of six weeks.
Students in the experimental group used the tapes during home practice and
were scored weekly on the performance of that weeks material. Poupolo
(1970) also used an extended treatment period of ten weeks. The students
practiced material to be learned in class but were tested on different music
that was sight-read by the student. None of the researchers except Poupolo
(1970) tested their subjects on how well the students transferred their

knowledge to new musical situations.

19

From an investigation of this literature, it is apparent that modeling is
an effective way to teach instrumental music to students at all levels of
development. There has been some evidence that the learning is generalized
to new musical situations (Poupolo, 1970). A review of this literature also
shows that the effects of modeling on the variable of tone quality has been
overlooked. It is the intent of this study to investigate the effects of modeling

on the development of beginning instrumentalists tone quality.

20

CHAPTER III
METHOD

This chapter presents a description of the research setting, the subjects
participating in the study, the research design, the procedures, and the
method used to collect and analyze data.

Subjects and Setting

Subjects that participated in this study were 65, fifth-grade beginning
band students in their first year of instruction. All of the students were
caucasian and were of diverse socioeconomic status. The sample included
boys and girls studying woodwind and brass instruments. The woodwind
instruments being studied were flute, saxophone, and clarinet. The brass
instruments being studied were trumpet, trombone, french horn, and
baritone. The students began instruction in September and worked for five
months with their band teacher before the study began in February.

Students’ regular band experience consisted of two 30-minute
rehearsals a week. Each 30 minutes was divided into sectional work and full
ensemble work The band was divided into a brass and percussion group and
a woodwind group for the first part of rehearsal and brought together for the
remainder of the rehearsal. The students were expected to practice at home in
traditional, self-directed sessions. Practice charts and rewards were not used to
reinforce good practice habits.

Procedure

This study was approved by the University Committee on Reasearch
Involving Human Subjects at Michigan State University (see Appendix A).
Letters of invitation to participate in the study were sent home to each

student (see Appendix B). The letter described the study and identified the

21

reasearcher as a graduate student pursuing a master’ 5 degree in music
education at Michigan State University. Each student in the study retumed a
letter signed by a parent showing consent for each child’s participation.

The researcher introduced the study to the subjects during band class
on a day previously agreed upon by the teacher and researcher. On that day,
the researcher introduced himself as a student at Michigan State University
who is pursuing a master’ 5 degree in music education. The researcher told the
students that he is conducting an experiment to find out what method of
home practice can help students like themselves improve more quickly on
their instruments.

All the subjects were randomly assigned a number and then assigned,
by number, to the experimental or control group. Students who took private
lessons were not included in the study. Their extra guidence may have
resulted in higher performance scores that would be attributable to their
private instruction and not the study’ 8 treatment. To establish group equality,
a pre-test was given. ”Ba Ba Black Sheep” and "This Old Man” were selected
from the Yamaha Band Student Book (Feldstein and 0' Reilly, 1988), as the
students had been rehearsing them in band rehearsals. Each student was pre-
tested by the researcher in a small study room that was prepared with a tape
recorder, microphone, stand, and chair. When the students arrived with their
instruments and books, they were instructed to turn to the appropriate pages
and study the selection. When the tape was started, the student stated his or
her number and played the selection. At the conclusion of each performance,
the tape was stopped and the student was thanked for his or her efforts.

The audio-taped performances were rated by two independent judges
who were knowledgeable in tone qualities of each instrument. The judges
were graduate students attending Michigan State University. One was the

researcher who was studying saxophone and music education and the other
was studying trombone and wind conducting. If the initial ratings for the
treatment and control groups were not statistically comparable, then the
analysis would have been adjusted using an analysis of covariance to
compensate for that difference. However, the results of the pre-test analysis
showed no significant differences between treatment and control groups for
tone quality or phrasing, so an analysis of covariance was not necessary.

All subjects participated in their regular band programs directed by
their band teachers during the 12 weeks of the study. The subjects in the
treatment group received listening tapes in addition to their regular band
experience. The listening tapes were to be incorporated in the daily practice
routine of the subjects. The subjects were instructed to warm up their
instrument, listen to the selections on the tape, and then begin their regular
practice session. The listening tapes included folk songs played by professional
musicians on the same instruments as the students. For example, the flute
players’ tape had selections played by a professional flutist and the saxophone
players’ tape had selections played by a professional saxophonist. The folk
songs were in major and minor tonalities and in duple and triple meters. The
selections were taken from the individual instrument tapes used with the
lump Right In: The Instrumental Curriculum. Each week the students had
two to three examples to listen to as part of their practice routine. At the end
of the week, the students exchanged the tapes for another tape containing the
next assigned listening examples.

Practice records were included on the tape jackets. Seven lines
provided represented a line for each day of the week. The students were to
record the date they practiced, the amount of time spent practicing, and if they
listened to the tape. A parent’s signature was required on the last line of the

23

tape’s jacket to verify that the infomation was accurately noted. At the end of
each week when the subjects exchanged their tapes, they received a colorful
pencil if they had returned the tape, recorded their practice time, and had a
parent sign the practice record. Practice records were only given to the
treatment group students since they were being used to insure the use of the
listening tapes during their already established home pratice routine.

After the 12 weeks of the study were completed, the students were post-
tested using ”Cherries Ripe” from the student’ 3 Yamaha Band Method book
(Feldstein and 0’ Reilly 1988). Subjects were given this piece two weeks before
the post-test so that they would have ample time to prepare. Having a
prepared piece helped to ensure that the study was measuring the student’ 8
ability to produce a quality tone and play with musically appropriate phrasing,
not the student’s ability to sight read. Some students may have a good sound
and stylistic sense when playing prepared music but not when sight reading,
especially at this age and experience level.

The students tape recorded their performance of this prepared piece in
the same manner as the pre-test. Due to a questionable inter-judge reliability
when the pre-test rating scales were used, they were altered for use with the
post-test performances. The inter-judge reliability was improved with these
new rating scales. Both sets of rating scales are found in appendices C, D, and
E, and the post-test rating scales are described in the next section. Two judges,
different from those who rated the pre—test performances, rated the post-test
tapes using the new tone quality and phrasing rating-scales. Both judges were
graduate students at Michigan State University. One was studying clarinet
performance and the other was studying saxophone and music education.

The researcher then analyzed the scores for significant differences.

24

Rating scales

The post-test rating scale for tone quality, which was designed by the
researcher, was based on a five-point Lickert scale. Robinson and Middleton
(1975) state that the production of a good tone is based on a student’ 3 aural
memory of a good tone. They go on the say that tone is the synthesis of
embouchure and breath support. Kohut (1973) concurs by saying that there is
no doubt that both embouchure and breath support are essential to good tone
quality. So, as students learn to make sounds on their instruments they must
master these two skills. First, they must learn to focus the embouchure to
control the vibration of either their lips for brass players or the reed for
woodwind players. Flute players focus their embouchure to control the shape
and size of the air stream. Second, they must learn to support the sound with
the appropriate amount of air for their instrument and the particular note to
be played. The rating scale then was designed to reflect the fact that both skills
play a part in tone development.

Embouchure development or focus and breath support or fullness
were rated separately on a scale of 1 to 5. A ”1” represented an unfocused
sound on the focus portion of the scale or an unsupported (thin) sound on
the fullness portion of the scale. A rating of "5” represented a focused sound
on the focus portion of the scale and a well-supported (full) sound on the
fullness portion of the scale. The focus score and the fullness score were then

averaged to attain a final tone quality rating.

Figure 3.1 Tone Quality Rating Scale

1 2 3 4 5
unfocused focused
average
1 2 3 4 5
thin full

The phrasing ability of the subjects was determined by listening for
how they connected their notes into phrases to create complete musical
statements. If the student played the selection in 4 bar phrases, as was
appropriate for the post-test selection, they were judged to have a good sense {
of phrasing. More specifically, phrasing sense was judged by awarding a point
for each phrasing opportunity in the selection. One point was awarded for
each measure that was played in a musically appropriate manner. This means
there were no breaks in tempo and the notes were connected enough to
present a single musical idea. Another point was awarded for each barline
that was included in the subject’ 3 phrase. Every barline was not included in
the scoring, because there are the standard phrasing points at which the
subject is expected to breath or pause. For the post-test selection, there were 28
possible points. Shown below is one of four lines of the post-test music with
point labels to show where the judges listened for connected musical ideas.

26

Figure 3.2 Point assignments for first phrase of post-test selection

I point 100““ I point

/\/\

I point
/\

t t i t

lpoint I point lpoint

 
 
 
 

   

 

no point awarded
(expected phrase
break)

The judges used a tally sheet and a score of ”Cherries Ripe” as they judged
each recording (see appendix E for tally sheet).

The judges were trained before each rating session. Five recorded
performances were chosen at random from the performance tapes collected by
the researcher. Each example was listened to, rated separately, and then talked
about to arrive at consensus on how to use the rating scales. When all felt
comfortable with how the rating scales were to be interpreted, the judging
session began.

Analysis of the Data

When all the data had been collected, the interjudge reliability was
determined between the two judges for each of the rating scales.
Intercorrelations between rating scales were determined. A two-tailed T-test
was used to analyse the data from each rating scale to determine if there was a
significant difference between treatment and control groups. If the scores for
the treatment group were significantly higher in phrasing, tone quality or
both, it would show that listening to a tape recorded model as a regular part of

27

practice is more effective than practicing with no recorded model in helping

students develop tone quality, phrasing or both.

28

Chapter IV
RESULTS

This investigation is concerned with whether a practice situation that
includes listening to tape-recorded models improves a beginning
instrumentalists tone quality and melodic phrasing as compared to students
who do not practice with such listening tapes.

The following statistics were calculated using the collected data: 1)
interjudge reliability of judges ratings for the tone quality and phrasing rating
scales, 2) means and standard deviations of the rating scale raw scores from
the treatment and control groups and, 3) t-tests to look for significant
differences between treatment and control group mean scores on phrasing
and tone quality.
lnteg'udge reliability results

Post-test interjudge reliability between the two independent judges was
determined for each rating scale (phrasing and tone quality) using the Pearson
R correlation. For the tone quality rating scale, reliability was found to be .787
and for the phrasing rating scale, reliability was found to be .778. This
represents a satisfactory amount of agreement between the judges’ ratings.
The intercorrelation between the phrasing and tone quality dimensions was
.55. The judges scores were combined for the following calculations.

Means and Standard Deviations

Table 4.1 shows the mean scores and standard deviations for the
treatment group and the control group for both tonal and phrasing
dimensions of the study. For the tone quality dimension, the treatment
group’s mean score is 6.13 and the control group’s mean score is 6.29. The
standard deviation for the treatment group is 1.98 and the standard deviation
for the control group is 2.15. For the phrasing dimension, the treatment

29

group’s mean score is 27 and the control group’s mean score is 27.43. The

standard deviation for the treatment group scores is 9.54 and the standard

deviation for the control group scores is 12.51.

 

 

 

 

 

 

 

 

 

 

Table 4.1
tone uality phrasing
7 mean std deviation mean std deviation
I control n=28 6.29 2.16 27.43 12.51
[treatment n=28 6.13 1.98 27.00 9.54
Resglts for tone gtLality

 

A two-tailed t-test was used to determine if there was a significant

difference between treatment and control mean scores for the tone quality

rating scale. The mean difference is .16 with 54 degrees of freedom (DF). The t-

value is .29 with a p-value of .77. Since the p—value is not less than .05, the

established level of confidence, the results are not statistically significant.

Table 4.2 shows the t-test results for the tone quality mean scores.

Table 4.2

Results of t-test on tone quality mean scores

 

[ Mean Difference l

l .16

Results for phrasing
A two-tailed t-test was used to determine if there was a significant

DF
54

I

J

t-value

I p-value

difference between treatment and control mean scores for the phrasing rating

scale. The difference in the treatment and control mean scores was .43 with 54

30

degrees of freedom (DF). The t-value was .14 with a p-value .89. Since the p-
value is not less than .05, the established level of confidence, the results are
not statistically significant. Table 4.3 shows the t- test results for the phrasing

mean scores.
Table 4.3

Results of t-test on phrasing mean scores

 

Mean Difference DF t-value p-value

 

 

.43 54 .14 .89

 

 

 

 

Interpretation of resfiu_lt_s_

The results show no significant difference between the scores of the
treatment and control groups in relation to tone quality or phrasing
development. The mean scores for control group were even slightly higher
than the mean scores for the treatment group. This outcome may have been
due to several factors.

The beginning of the study fell several months into the student’ 5
instruction. This may have allowed poor playing habits and a poor aural
concept of tone and phrasing to develop before the study began. These poor
playing habits and / or aural concepts may have resulted in 12 weeks being an
inadequate amount of time for observable improvement to manifest itself.
Students would need to unlearn those poor playing habits and aural concepts
and replace them with new ones.

At the time of the study, the band program had no established system
holding students accountable for their home practice habits. It was assumed,
though, that the students were practicing to some degree, so there would be
only a need to monitor the treatment group’s use of the listening tapes. The

31

failure of a significant difference in the treatment group scores may have been
due to a lack of practice all together. The students who were self-motivated
enough to practice and did, may have been concentrated in the control group,
resulting in the group’s higher mean scores.

Based on the practicing assumption, practice records and incentives
were only used with the treatment group. This was to insure that the students
added the listening tape to their home practice session routine. The reward of
a pencil for practicing and returning the tape may not have been incentive
enough to spark a desire to practice. In fact many of the tapes were returned
with few or no practice sessions recorded. As a result, it was difficult to assess
whether students were just not practicing or forgetting to notate their practice
times.

It is possible that the students may have experienced nervousness
during the performance of their post-test selections, causing them to perform
poorly. All effort was made by the researcher to ease the nerves through
encouragement and warm up opportunities. Still, the demands made by the
school schedule may have caused the researcher and teacher to hurry
students through the process, affecting student nerves, concentration and
performance accuracy.

It is possible that the students did not have access to a quality tape
player. When asked by the researcher and teacher if they had access to a tape
player, all treatment group students indicated that they did. However, the
tape-players owned by the students may have been of poor quality. This
would present a poor example of tone quality and possibly hide the subtleties
of phrasing that quality equipment would reveal.

Although music aptitude was not considered as a variable in this study,
high aptitude students may have absorbed the information from their teacher

32

more quickly, making faster gains than the lower aptitude students. It is
possible that high aptitude students were concentrated in the control group
causing the group’s higher mean scores. Aptitude was not measured at the
beginning of the study so no data exists for review on this matter.

At the beginning stages of learning to play an instrument, students are
learning many things at once; among other things, they have to learn how to
blow, to finger, and to read music. It is possible that this study was measuring
the students ability to read music rather than the student’s ability to play
musically with a good sound. The task of reading the notation may have
gotten in the way of the student’s audiation of the tune and production of the
desired tone quality. Stopping to decode a note on the staff could definately
get in the way of performing a flowing phrase.

It is possibile that the student’ 3 embouchure and diaphramatic control
are simply not developed enough to show in performance what he/ she may
have gained aurally. A year or more may be required for the student to have
control enough over these faculties to express what they can audiate.

The interjudge reliability was deemed acceptable, yet it was still
somewhat lower than interjudge reliabilities found in previous studies. It is
possible that the judge preparation could have been longer and more in-
depth. Due to the limited availability of the selected judges, time was short.
The judges practiced and discussed answers on 5 practice examples before
beginning their independent judging. This may have led to greater
subjectivity in scoring. This subjectivity in scoring could have affected the
agreement between judges scores reducing the interjudge reliability.

The study relied heavily on the teaching of the band director, since the
it was designed to test a practice supplement to regular instrumental
instruction. The students were reminded by the researcher to be thinking of

what they had learned in their lessons and band classes while they practiced.
In order to produce tone and phrasing like they heard on the tape, they had to
employ learning and skills gained from their teachers during band classes. If
the instruction that the tapes were supposed to be aiding was poor or
inconsistent, the students may not have been gaining the skills needed to
achieve the desired tone or phrasing. This was a concern of the researcher
during the study. There were two teachers involved in the instruction
process. The brass and the woodwind teachers did not teach with the same
level of excellence and attention with respect to posture, hand position,
embouchure formation and breath support. It appeared that the woodwind
teacher was not as well liked as the other teacher. This lack of rapport may
have affected the motivation of the students to work hard, affecting their
practicing habits and, as a result, the results of the study suffered. The two
teachers did not get along, and both had conflicts with the administration of
the elementary school. The inner struggle of the music department may have
contributed to an atmosphere non-conducive to such a study that relyed so
heavily on a successful band program.

Though the results showed nothing of significance, factors such as
practice habits, timing of the study, music aptitude, grouping of the students,
lack of quality play back devices, judging subjectivity, and a tense learning
environment may have played a role in these results. However, it is also
possible that spending time listening to musical examples that are not being
currently studied and performed by outstanding models as part of a practice
routine may have no effect on the tone quality and phrasing of beginning

instrumentalists.

CHAPTER V
SUMMARY, CONCLUSIONS, RECOMMENDATIONS, AND
IMPLICATIONS

The purpose of this study was to investigate the effects of aural models
on the development of elementary student’ 5 performance skills. There are
two specific problems of this study. First, will students develop a more
characteristic tone on their instrument if, during home practice, they listen to
an aural model of a professional playing the same instrument they are
studying? Second, will students develop a better sense of phrasing (style) if,
during home practice, they listen to an aural model of a professional playing
the same instrument that they are studying?

Summary

The study was set up in an treatment/ control group design.
Approximately 56 fifth-grade instrumentalists from a rural Michigan
elementary school participated in the study. Only students who were studying
wind instruments were included in the study. The students were taught by
their instrumental teacher in their regular band program throughout the 12
weeks of the study, because the study was looking at the effectiveness of
materials that would supplement the regular band experience of a student.
Each student was randomly assigned to one of the two groups, treatment or
control, and a pre-test was given to establish equality of the groups.

Students in the treatment group were given an audio cassette tape of
professionals playing folk songs on the same instruments they were learning.
Each week, a new tape was exchanged for the previous week’s tape. Each tape
consisted of two or three listening examples that the students were to listen to

at the beginning of their home practice sessions. The control group carried
out non-supplemented home practice sessions. To encourage students in the
treatment group to use the listening tapes, a practice record was included with
each tape. The student recorded the date, the amount of time practiced, and if
they listened to the tape. Upon exchanging the tape, the student was rewarded
with a colorful pencil if they listened and practiced.

At the end of the 12 weeks, a piece from the students band method
book was chosen by the researcher and the teacher to serve as a post-test. Each
student individually performed the post-test selection for the researcher, and
the performances were audio-taped.

The post-test performances were rated using two separate rating scales
by two independent judges, neither of whom were the researcher. The tone
quality scale had two elements, focus of sound and fullness, that were scored
separately and then averaged. The phrasing scale consisted of a tally of points.
A point was awarded for each phrasing opportunity; a phrase opportunity
could be performed within a measure or over a barline excluding the barline
at the end of every four measures. The judges were trained by the researcher
before rating the actual performances. This was to ensure that they listened
with the same understanding of the elements being measured.

After the student post-test performances were rated, the data was
analyzed for reliability between the two judges and significant differences
between the treatment and control groups. The means and standard
deviations were also calculated for the post-test scores for the treatment and
control groups.

Conclusions

The following conclusions can be considered valid for only the

population in this study. The treatment and control groups showed no

significant difference between one another in their development of tone
quality or phrasing. These results are in direct conflict with the body of
research on modeling. Modeling has been shown effective in improving
performance skills by Rosenthal et. al. (1988) and Poupolo (1970) in specific
relation to phrasing or expression and in other performance areas such as
intonation, rhythmic accuracy, pitch accuracy, and tempo by Rosenthal (1984),
Zucher (1975), and Fortney (1992). Since tone quality development was not
addressed by the above research, and this study found no significant results,
more research needs to be done in this area to support or negate the results of
this study. It may be that the material on home practice tapes needs to be
material that the student is studying and playing to have any positive effect
on tone quality and phrasing development.
Recommendations for Future Research

The following recommendations based on this experiment are
concerned with the need to improve the procedures used in this study. Also
included are suggestions for further studies on the effects of aural models on
the performance skills of phrasing and tone quality.

The fact that the results of this study conflict with other research points
to the need for additional research. The study should be replicated with a
more effective practice incentive and accountability system. Getting the
students to practice what is assigned is essential to discovering possible effects
of a modeling treatment. It is also recommended that a school system be
chosen that has an established system of student practice accountability in
place.

Since the study began well into the school year, the lack of practice
accountability may have fostered poor practice habits in this circumstance.
Because of this, it is recommended that this study be replicated starting at the

37

beginning of the school year. This way, the practice ethic and the use of the
tapes can be taught from the start, helping to gain valid results.

Students may have been hindered in their tone quality and phrasing
development due to their struggle with notational issues. Stopping to decode
notes on the staff will certainly interupt breath support and relax the
embouchure affecting the students’ tone quality and flow needed for good
phrasing. It is recommended that this study be replicated using rote songs .
without notation as the pre- and post-test selections. The students could
possibly be taught one or two songs by ear prior to the pre-test and then
divided into the treatment and control groups. By the end of the study they
would have many songs that they can play by ear in their repertoire. The
effects of a supplimental tape may be easier to observe, since students would
not be working to read notation.

This study’ 5 results did not negate the findings of studies that showed
the effectiveness of home practice tapes where studied material is modeled.
This study could also be replicated using performances by excellent models of
the material being learned by the student. This would be a more direct model
of the tone quality and phrasing desired by the teacher and possibly transfer
with greater impact to other music they might perform.

Music aptitude was not considered in this study. It would be interesting
to see whether exposure to aural models would affect students with high
aptitude differently than those with low or medium music aptitude. Such a
study could also take into consideration the students current skill level upon
starting the study.

Replication of the studies presented in chapter four would be
recommended. This would bring the current body of research up to date and
keep strengthening the findings of the past 30 years.

It would be interesting to conduct a study concerning aural concepts of
music that students use to learn songs on their instruments. If students were
required to be able to sing their songs accurately before they could play them
on their instruments, would they play them better and be able to deal with
reading notation more easily?

Implications

Existing literature supports that modeling positively affects the
development of performance skills in beginning instrumentalists. Although
the present study does not support this literature, it is possible that further
research focusing on tone quality and phrasing will find supporting results.
Teachers should experiment formally and informally with new ways to
convey musical information through the use of aural models. By doing this,
new ideas can inspire new questions. These new questions can lead to new
studies that would add to the body of knowledge available to music educators

who seek to improve their student’s learning.

39

APPENDICES

4O

 

OFHCEOF
RESEARCH
AND
GRADUATE
STUDIES

University Committee on
Research Involving
Human Subjects
(ucnmS)

Michigan State University
246 Administration Building
East Lansing, Michigan
4&QtuM6

517/355-2180
FAX" 517/432-1171

fee Mcrugan Slate Universuy
IDEA IS Institutional Diversity
fry-g were in Minn

Appendix A
PILHVLAIJ SIIBTELITSIFCDRUVI

h4|€3ti|(3%\li Sillilli

 

UNIV

February 13,

Siltfiiiigagiééi Dr

TO:

RE: IRB#:

TITLE:

ERSITY

1997

97-035

AN INVESTIGATION OF THE EFFECTS OF AURAL MODELS
OF THE DEVELOPMENT OF BEGINNING
INSTRUMENTALIST'S TONE PRODUCTION AND STYLISTIC

CONCEPT
REVISION REQUESTED: N/A
CATEGORY : l -A
APPROVAL 02/10/97

The.University
revrew of this
rights and welfare 9. the human subjects appear to

DATE :

Committee on Research Involving Human Subjects'(UCRIHS)
project is complete. I am pleased to advrse that the
f be adequately

rotected and methods to obtain informed consent are appropriate.

herefore,
above.

RENEWAL :

REVISIONS :

PROBLEMS /
CHANGES :

If we can be of any future hel
at (517)355-2180 or FAX

Sincerely,

  
   

DEW:bed

CC:

D vid E. Wright“
CRIHS Chair

the UCRIHS approved this project and any revrsions listed

UCRIHS approval is valid for one calendar year, beginning with
the approval date shown above. Investigators planning to
continue a project beyond one year must use the green renewal
form (enclosed with t e original approval letter or when a.
project is renewed) to seek u date certificatron. There_1s a
maxrmum of four.such expedite renewals ossrble. Investigators
wishing to continue a progect beyond tha time need to submit it
again or complete revrew.

UCRIHS must review any changes in procedures involving human
subjects, rror to initiation of t e change. f thrs rs done at
the_trme o renewal, please use the green renewal form. To
revrse an approved protocol at any other time during the year,
send your written request to the. CRIHS Chair, requesting revised
approval and referencrng the progect's IRB # and title. Include
in our request a description of the change and any revised

ins ruments, consent forms or advertisements that are applicable.

Should.either of the followin arise during the course of the
work, investigators must notr y UCRIHS promptly: (1) roblems
(unexpected srde effects, comp arnts, etc.) involving uman
subjects.or (2).changes 1n the research environment or new
information indicating greater risk to the human sub'ects than

existed when the protocol was previously reviewed an approved.

lease do not hesitate to contact us

(517)4gé- 171.

 
  

Peter O. Conniff

41

Appendix B
CONSENT LETTER SENT TO PARENTS
Monday, January
27, 1997
Dear Parents,

I am a graduate student at Michigan State University pursuing a
master’ 5 degree in music education. In partial fulfillment of this degree, I
have chosen to do a research study involving beginning instrumental
students. Your school was selected to participate in this study. The following
is a description of the study and a consent form giving permission for your
child to participate in the study. Participation will not in any way affect the
student’ 3 band grade and is strictly voluntary.

The study will take the form of an experimental design and last for 12
weeks. This means that there will be two groups of students, both receiving
regular band instruction but the home practice technique will be different.
The second group will borrow a different listening tape each week to be used
during home practice. The results of the study will show if this addition to
the home practice routine will improve students playing in the areas of tone
quality and stylistic understanding.

The student performances will be rated by two judges twice during the
study. The identity of student’ 3 performances will remain confidential. If
there is no objection to your child’s participation in the study, please sign the
consent form below and return the bottom portion to Ms. Berry by Friday,
January 23, 1997. Keep the top half of this letter for your information. Thank
you for your support.

Sincerely,

Peter Conniff

Consent form
[ ]Yes, I give permission for my child to participate in the study
[ ] No, I do not give permission for my child to participate in the study

Parent or Guardian Signature: Date

43

Appendix C
PRE-TEST RATING SCALES

PRE-TEST RATING SCALE FOR TONE QUALITY

1 produces a sound

2 produces discrete pitches
3 thin , unfocused sound

4 full , but unfocused sound
5 full, focused sound

PRE-TEST RATING SCALE FOR PHRASING

1 Mostly plays note to note with no smooth connections
2 Mostly connects notes in one measure phrases

3 Mostly connects notes in two measure phrases

4 Mostly connects notes in three measure phrases

5 Mostly connects notes in four measure phrases

Appendix D
POST-TEST RATING SCALE SHEET

1) focus 1 2 3 4 5
average

fullness 1 2 3 4 5

2) focus 1 2 3 4 5
average

fullness 1 2 3 4 5

3) focus 1 2 3 4 5
average

fullness 1 2 3 4 S

5) focus 1 2 3 4 5
average

fullness 1 2 3 4 5

6) focus 1 2 3 4 5
average

fullness 1 2 3 4 5

7) focus 1 2 3 4 5
average

fullness 1 2 3 4 5

8) focus 1 2 3 4 5
average

fullness 1 2 3 4 5

9) focus 1 2 3 4 5
average

fullness 1 2 3 4 5

10) focus 1 2 3 4 5
average

fullness 1 2 3 4 5

11) focus 1 2 3 4 5
average

fullness 1 2 3 4 5

45

Appendix E
POST-TEST PHRASING TALLY SHEET

LIST OF REFERENCES

47

LIST OF REFERENCES

Anderson, J. N. (1981). Effects of tape-recorded aural models on sight-
reading and performance skills. Journal of Research in Music Education,
22(1), 23-30.

Baker D. S. (1980). The effect of appropriate and inappropriate in-class
song performance models on performance preference of third- and fourth-
grade students. Journal of Research in Music Education, 28, 2—17.

Brand, M. (1977, Summer). Effectiveness of simulation techniques in
teaching behavior management. Igurnal of Research in Music Education,
_2_§(2), 131-138.

Davidson, L (1989). Observing a yang ch'in lesson: learning by
modeling and metaphor. Lqqunal of Aesthetic Education 23(1), 85-99.

 

Forsythe, J., and Gonzo, C. (1976). Developing and using videotapes to
teach rehearsal techniques and principles. Journ_al of Research in Music
Education, 24 , (1), 32-41.

Fortney, P. M. (1992). The effect of modeling and silent analysis on the
performance effectiveness of advanced elementary instrumentalists. Research
Perspectives in Music Education, 3, 18-21.

Gordon, B. (1997). Learning fluences In Music. (Rev. ed.). Chicago:
GIA.

Guralnik, D. B. (Ed.). (1984). Webster’s New World Diegonary of t_h_e_
American Langu_ag_e_. New York: Warner Books.

Jetter, J. T. (1978, Summer). An instructional model for teaching
identification and naming of music phenomena to preschool children.

Journal of Research in Mgsic Education, 26,(2), 97-110.

48

Kohut, D. (1973). Instrumental music pedagogy: teaching techniques for
bani and orchestra directors , (pp. 99). Englewood Cliffs, NJ: Prentice Hall.

Leonard, C., 6: House, R. W. (1972). Foundations and prinmpals of
music education (2nd ed.). New York: McGraw-Hill.

McCarthy, J. F. (1974). The effect of individualized instruction on
performance achievement of beginning instrumentalists. Council for
Research in Music Education, 38, 1-16.

Middleton, J. A., Robinson, W. C. (1975). Complete school band
program, (pp. 72). San Antonio, Tx: Southern Music.

Moore, R. S. (1976). Effect of differential teaching techniques on
achievement, attitude, and teaching skills. Mal of research in Music
Education, 2_4(3), 129-141.

Perry, J. C., Perry, I. W. (1986). Effects of exposure of classical music on
musical preferences of preschool children. Mal of Research in Music
Education, 34, (1), 24-33.

Poupolo, V. (1970). The developmerLand experimental application of
self-instructional practice materials for beginning instrumentalists.
Unpublished doctoral thesis, Michigan State University, East Lansing.

Rosenthal, R. K. (1984). The relative effects of guided model, model
only, guide only, and practice only treatments on the accuracy of advanced
instrumentalists performance. lournal of Research in Music Education, 32,
275-273.

Rosenthal, R. K, Wilson, M., Evans, M., 81: Greenwalt, L. (1988). Effects
of different practice conditions on advanced instrumentalists’ performance.
lggmal of Research in Music Edpcation, 36, 250-257.

Zucher, W. (1975). The effects of model-supportive practice on
beginning brass instrumentalists. In C. K Madsen, R. D. Greer, 6r C. H.

49

Madsen, Jr. (eds), Research in music behavior: Modifyinginusic behavior in

me glassroom (pp. 131-138). New York: Teachers College Press.