Journal of Research in Music Education

Patricia Shehan Campbell is the recipient of the MENC 2002 Senior Researcher Award. The following speech was presented on April 12, 2002, at a special session of the Society for Research in Music Education at MENC's National Biennial In-Service Conference in Nashville, Tennessee.

Forty-one sighted and 17 blind children listened individually to six short music excerpts and described them orally. Each child's descriptions were recorded, transcribed, and randomly ordered. Then, a panel of 10 music teachers attempted to assign each description to its intended excerpt. Teachers were able to match description to excerpt with about the same degree of accuracy for blind and sighted children. Descriptions were more easily matched at successive grade levels. Analysis of types of language used showed that sighted children remained consistent in number of musical elements described across age-groups; however, blind students increased substantially in their descriptions of musical elements at each successive level (kindergarten, primary, upper elementary). Sighted children used significantly more metaphors and emotional descriptors than did blind children. The use of temporal language increased with age, particularly among the blind students, but there were no statistical differences due to visual ability in the present study.

The purpose of the current study was threefold: (a) to determine the nature of junior high school instrumental music students' self evaluation tendencies over time, (b) to examine whether the process of self evaluation, with or without the use of a model, has an effect on self-evaluation accuracy, and (c) to determine if a relationship exists between self evaluation accuracy and music performance achievement. Forty-one junior high woodwind (h = 28), brass (n = 10), and percussion (n = 3) students participated in the pretest/posttest 2×2 factorial design. Data indicated that students' self-evaluation scores increased and their self-evaluation scores did not improve over time regardless of model-group condition. Furthermore, intonation accuracy may deteriorate. A Pearson correlation showed no correlation between expert and student evaluations in all pretest performance subareas and most posttest subareas. There was a moderate positive relationship between model-group tempo and combined group interpretation posttest scores.

Since the advent of children's educational television, there has been much speculation but little research conducted on the possible effect of such programming on children's music learning. The present investigation was undertaken to determine if third-grade children could identify songs from the show Barney & Friends. Results indicated that (a) there was not a positive relationship between students' self-reported frequency of watching Barney & Friends and song recognition; (b) taken as a whole, the group of popular songs was identified significantly more often than not, whereas the group of “Barney ” songs tended not to be recognized by the group of frequent viewers; (c) when songs had their lyrical content altered on the show, students did not tend to remember those songs in the new formats; and (d) boys tended to say that they watched the show significantly less often than did girls.

The purpose of this study was to develop a valid and reliable assessment measure for stringed instrument performance. In the initial phase of the investigation, a total of 90 suitable statements were generated for the initial item pool gathered from essays, statements, and previously constructed rating scales. These statements were put into the a priori categories determined by previous research. These items were paired with Likert-type scales and used by 50 judges to assess 100 recorded string performances at the middle school through high school level. The results of the initial item pool were factor-analyzed using a varimax rotation. Five factors were identified (interpretation/musical effect, articulation/tone, intonation, rhythm/tempo, and vibrato), and 28 items were selected for the subscales of the String Performance Rating Scale (SPRS) based on factor loadings. Reliability varied from .873 to .936 for each judging panel using Hoyt's analysis of variance procedure. Two studies were conducted to establish criterion related validity, with zero-order correlations ranging from .605 to .766 between the SPRS and two other rating scales.

Research has shown that undergraduate students' self-evaluation of performance correlates poorly with instructor and peer evaluation. This article reports two exploratory investigations into the development of a treatment condition for improving performance self-evaluation. The condition consisted of small groups of peers informally discussing performances and sharing feedback with one another.

The first investigation resulted in a statistically significant difference between experimental and control groups in ability to self-evaluate, although the effect size was small. With a second investigation, we pursued a modified version of the treatment emphasizing changes over time in ability to self-evaluate. In the second investigation, we also examined different effects of this modified treatment condition on students whose initial attempt to self-evaluate was either accurate or inaccurate. The second investigation did not result in significant differences between treatment and control groups; however, a significant interaction between time (self-evaluation across five small-group peer-interaction sessions) and initial ability to self-evaluate accurately was noted. A prompt improvement was found with performers whose initial ability to self-evaluate accurately was poor, although the effect tended to fade over time.

Consistent with prior research, self-evaluation did not correlate highly with instructor evaluation. Also consistent with prior research, peer evaluation was higher than instructor evaluation. Correlations between instructor evaluation and peer evaluation declined over the five sessions. Self-evaluation scores increased over time, moving away from instructor evaluation scores and toward the higher peer-evaluation scores.