The Effect of Three Evaluation Procedures
on the Rehearsal Achievement of
Eighth-grade Band Students

Dr. William Zurcher

In Applications of Research in Music Behavior: The Alabama Project (Music, Society, and Education in America). Alabama Press, © 1987, 51-58. The University of Alabama, Tuscaloosa, AL, and London, England. Edited by Clifford K. Madsen and Carol A. Prickett, with a foreword by J. Terry Gates. ISBN 0-8173-0305-7. (See book review in Music Educators Journal, February 1988, 19-20.)


Achievement was significantly higher (p < .05) when students recorded their own daily numerical grades (A) than when they received no feedback (B). Daily numerical grades (A) also resulted in significantly higher achievement (p < .001) than daily teacher-issued grades.

This study indicates that daily teacher-issued letter grades are not specific enough to change behavior. Because so many grading systems utilize letter grades, further research is urged in music and other subject areas to structure task-specific measuring systems.

Design: Withdrawal (ABCA) N = 63 Groups: 1

Statistical Analysis: Correlated t tests (two-tailed), Pearson product-moment

Interobserver Reliability of Student Entries: .98

Independent Variables: A. Student-recorded/teacher verified daily numerical grades; B. No feedback; C. Teacher-issued daily letter grades.

Dependent Variable: Grade means (0-100)

Music: Performance, group


The Effect of Report Card Frequency
On the Performance Achievement of
Junior High Instrumentalists

Dr. William Zurcher

Paper presented at the Sixth National Symposium on
Research in Music Behavior, Fort Worth, 1985


This study was designed to determine whether differences in achievement exist between students who receive a take-home report card every six weeks and students who receive a take-home report card every twelve weeks. The eighth-grade band at Baldwin Junior High School was randomly divided into an experimental group of 31 and a control group of 31. At the end of a twelve-week baseline period, both groups received grade and attitude marks on their school report cards. The conrol group continued to receive marks every twelve weeks and the experimental group was informed that they would receive marks every six weeks.

Analysis of t-test values showed no significant differences in either grade or attitude marks between the two groups.


A Credit System for the Baldwin Junior High School Band

Dr. William Zurcher

In R. D. Greer, Design for Music Learning. (Follow link and see bottom of first page for reference to book.) New York and London: Teachers College Press, © 1980, 150-157. ISBN 0-8077-2573-0.

NOTE: The following synopsis was issued to band members by the author and is contained in the book along with several forms designed by the author.

“The following credit system shall serve as a basis for all band grades and attitude marks. This system offers a positive approach to grading. Rather than punishing the student for musical and social errors through demerits, the credit system rewards appropriate responses and behaviors. It also provides a measurable index of improvement in at least 25 areas. Because students are responsible for recording their own credits, they tend to be more aware of what constitutes appropriate musical and social behavior.

In addition, the enrichment nature of the credit system is important. Students are motivated to earn credits since they may be spent. Credits may be earned in rehearsals and lessons and for completing modules in harmonic dictation, improvisation, form, composition, guitar, teaching or observing, and for tape-recording solo accompaniments, band parts, or compositions. These earned credits may in turn be spent for additionally enriching music reinforcers or rewards such as conducting, leading improvisations, improvising with the band or jazz/rock ensemble, receiving lessons on a second instrument (from other students), playing a second instrument during free time, lunch or band (contingent on being able to play band parts), borrowing a tape-recorder overnight, taking a second instrument home overnight, checking out solo or ensemble music or working on personal music projects during a band rehearsal or lesson period (if certain conditions are met). Credits may also be spent for certain non-music rewards such as coming late to rehearsal, leaving early, and staying in the music area during free time, or before or after school. NOTE: CREDITS SPENT DO NOT AFFECT GRADE OR ATTITUDE MARKS IN ANY WAY.

Adjustments in the credit system may occasionally be necessary. These changes will be posted on the bulletin board and announced to the band at least two weeks prior to the effective date of the change.

Each marking period the grade and attitude mark will be based on the percentage of credits earned from the total possible for that marking period.”

NOTE: A number of forms are contained in the book.


The Overjustification Hypothesis and
Choice of High and Low Preference Music and Non-Music Activities
Over an Extended Time Period in a Junior High Band Program

Dr. William Zurcher and Dr. R. Douglas Greer

Paper presented at the Fourth National Symposium on
Research in Music Behavior, New York, 1980


A quasi withdrawal design (A, B, A1 , B1, A) was used to study the effectiveness of procedures used to enhance students' election of comprehensive music learning projects. The procedures were evaluated within the confines of a Personalized Systematic Instruction (PSI) program for an 80-member junior high band. During the first enhancement phase (B), a single verbal description of the value of spending earned credits (tokens) for music learning projects was given during a six-week grading period. During the second enhancement phase, (B1), daily verbal descriptions and demonstrations of projects were available. Data showed that during the initial baseline phase, no projects were elected. After the initial baseline, students elected to spend credits (earned for performance and deport­ment) on music learning projects. More projects were elected during enhancement phases (B and B1 ) than during baseline phases (A, A1, and A) and the greatest number were elected during the B1phase.

The validity of the overjustification hypothesis for PSI music programs was tested by an ex post facto analysis of student choices of music programs versus avoidance of music activities. Students overwhelmingly elected to spend credits for music projects than for avoidance of music activities both in enhancement and no-enhancement phases. Data were collected over an eight-month period in the third year of a PSI program for this band.


A Personalized Systematic Instruction
Program for Band: Third Year Report

Dr. William Zurcher

Paper presented at the Third National Symposium on
Research in Music Behavior, Atlanta, 1978


This paper is a summary of data collected during a three-year period of Personalized Systematic Instruction (PSI) with the Baldwin Harbor Junior High School Band, Baldwin, New York. The grade average for three years was 89.32 and the attitude average was 90.94. Pearson Product-Moment correlations were .91 using student observers and .992 for teacher observations.

Students used self-recording procedures to keep track of their daily grade and attitude marks in 23 categories. During a six-week reliability check, students (utilizing printed tally-sheets) made 17,524 entries with only 132 errors (unearned credits).


The Effect of Model-Supportive Practice
on Beginning Brass Instrumentalists

Dr. William Zurcher

Doctoral Dissertation presented at the Music Educators National Conference, Philadelphia, 1975. Dissertation Abstracts International, 33, 5A.


It was the purpose of the study to investigate whether cassette-recorded models and instructions for home practice (model-supportive practice) would be more effective than traditional practice methods in improving performance achievement. A review of related literature indicated that (l) non-book programmed instruction and computer-assisted instruction for instrumentalists parallel the results found in other subject areas, (2) differences between prompting, confirming, or simultaneous stimuli can not be said to be conclusive, (3) it has not been shown that there are significant differences between, aural, visual, or verbal feedback for beginning instrumentalists, and (4) imitative tendencies are learned and established through reinforcement and can be transferred to other similar situations.


The basic experimental design utilized is described by Campbell and Stanley (l963) as a Posttest-Only Control Group Design. However, the design was extended in order to post test each weekly assignment over a period of six weeks. This permitted both weekly (vertical) and cumulative (horizontal) measurements of results.

Group A (N=2l) R XO CO XO CO XO CO
Group B (N=22) R CO XO CO XO CO XO

N = Number of subjects X = Experimental
R = Random assignment  C = Control
O = Observation

Forty-three beginning fourth, fifth, and sixth grade trumpet, horn, trombone, and baritone students from two elementary schools in Baldwin, New York, were randomly assigned to an experimental and a control group for the first week of practice. Each week thereafter the groups rotated treatments.

An original method book was used and an individual cassette tape and recorder were issued weekly to each experimental subject (S). Each tape included instructions, reminders, and model "play-along" performances of the music.

Each S received a weekly l5-minute individual lesson from the investigator during which post tests of assigned material were administered. Following each post test needed corrections were made and instructions given for the next lesson. New lesson booklets and tapes were distributed and previous booklets and tapes collected each week. Ss were not permitted to repeat lessons.

Normally distributed cumulative data (an experimental and a control score for each S) were analyzed with t tests for correlated data. Product-moment correlations were used to evaluate raw data reliability.

Reliability results on the collection of raw data are as follows:
1. Pitch Errors: r = .99
2. Tempo Stability: r = .99
3. Pitch Matching: r = .99
4. Fingering and Slide Errors: r = .98
5. Rhythm Errors: r = .99
6. Total Practice Time: r = .74


Six null hypotheses (which were required to meet at least a .05 level of significance on a two-tailed test in order to be rejected) were tested, resulting in the following conclusions:

  1. Model-supportive practice is more effective than traditional practice in reducing pitch errors exceeding +/- 100 cents.
  2. There is no difference between model-supportive practice and traditional practice in establishing tempo stability.
  3. Model-supportive practice is more effective than traditional practice in developing pitch-matching skills.
  4. There is no difference between model-supportive practice and traditional practice in reducing the number of fingering or slide position errors.
  5. Model-supportive practice is more effective than traditional practice in reducing rhythm errors.
  6. Model-supportive practice is more effective than traditional practice in increasing the time spent in practice.


Since a difference was shown between the two treatments in total practice time, the question must be raised of whether differences in performance achievement were caused by increased practice or by the model-supportive nature of the practice tapes. Spearman rank correlation coefficients showed no correlation whatever between total practice time and performance achievement. Performance differences which do exist may then be attributed to the model-supportive nature of the practice tapes, i.e., to the quality and not the quantity of practice.


Campbell, D. T., & Stanley, J. C. Experimental and quasi-experimental designs for research. Chicago; Rand McNally, 1963


The Effect of Model-Supportive Practice
on Beginning Brass Instrumentalists

Dr. William Zurcher

In Research in Music Behavior: Modifying Music Behavior in the Classroom. Teachers College Press, Teachers College, Columbia University © 1975, 131-138. Edited by Clifford K. Madsen, R. Douglas Greer and Charles H. Madsen, Jr. Library Of Congress Catalog Number: 71-16362. ISBN 0-8077-2436-X. This article is based on the author’s 1972 EdD dissertation of the same title at Teachers College, Columbia University.

Summary of the Study

. . . (This) study indicates that a take-home practice model does indeed improve performance and further corroborates the lack of a relationship between time spent practicing and achievement.

The effect of model-supportive practice seems applicable to most applied study. Giving the student something specific to do and something specific to listen for as opposed to just going through the music seems applicable across all students and situations. This study also point out the necessity for careful scrutiny of modeling effects in music learning.

Design: Post-test only/control group N = 43 Groups: 2

Statistical Analysis:t tests, Mann-Whitney U

Reliability: .74-.99

Independent Variables: Tape-recorded (take home) practice models

Dependent Variables: Fingering errors, gross pitch deviation, temporal stability, pitch accuracy

Music: Performance, Listening, Individual, Group


Back to top | Home

E-mail: KayZurcher@gmail.com