Measuring the Sources of Self-Efficacy Among Secondary School Music Students

Abstract

The purpose of this study was to investigate the four sources of self-efficacy in music performance and examine responses from the Music Performance Self-Efficacy Scale (MPSES). Participants (N = 290) were middle and high school music students from 10 schools in two regions of the United States. Questions included the following: (1) How much influence does each source have on self-efficacy? (2) Are there differences among band, chorus, and string students? (3) Are there differences between middle and high school students? (4) Does music aptitude predict self-efficacy, and (5) Does evidence support the MPSES as a valid and reliable scale? Results indicated that mastery experience exerted the strongest influence, no differences were found among ensemble types or grade levels, and music aptitude scores predicted modest increases in self-efficacy, β = .16 (.07). Examination of test content, response process, internal structure, and relationships with other variables provided evidence of validity, while internal consistency and test-retest values provided evidence of reliability. Recommendations for further research included examining the development of self-efficacy at different ages, measuring the relationships between the sources of self-efficacy and music achievement, and using the MPSES to assist in the investigations of other psychological constructs, such as self-identity and motivation.

Keywords

self-efficacy music performance music aptitude music education sources of self-efficacy

Students develop self-perceptions of their music abilities as they acquire knowledge and skills in the classroom. McPherson and McCormick (2000) stated “how students think about themselves, the task, and their performance is just as important as the time they devote to practicing their instrument” (p. 31). Studies have been conducted on the relationships between various self-perceptions and participation in music activities. Researchers have investigated self-concept (Austin & Vispoel, 1998; Greenberg, 1970; Svengalis, 1978), self-esteem (Randles, 2010; Schmitt, 1979), self-identity (Lum, 2008), and self-regulation (McPherson & Zimmerman, 2002; Miksza, 2011). Of particular interest are studies on self-efficacy in music performance. Findings from these studies have indicated that self-efficacy has a particularly strong relationship with achievement in music performance (McCormick & McPherson, 2003; McPherson & McCormick, 2000, 2006). In addition, Miksza (2011) identified self-efficacy as a theoretical precursor to indicators of intrinsic motivation, such as goal setting and persistence on task. Given these findings, further research into the self-perception of self-efficacy is warranted and may provide valuable insights for improving student achievement in music performance.

Researchers have used a variety of approaches to measure self-efficacy in music. They have developed their own questions to elicit self-efficacy beliefs (McCormick & McPherson, 2003; McPherson & McCormick, 2000, 2006), modified items in existing scales to reflect the subject of music (Martin, 2012; Nielsen, 2004; Ritchie & Williamon, 2011), and combined original questions with those from preexisting scales (Hendricks, 2009; Wehr-Flowers, 2007). In most studies, researchers measured self-efficacy as a single composite construct immediately before a performance. These summative assessments provide useful information for researchers but little information of value for educators in the classroom. In general, summative assessments are effective tools for identifying students with high or low levels of self-efficacy but offer no insight into the development of the construct. Music educators need a formative assessment that tracks the development of self-efficacy beliefs and that also can deliver feedback on the effectiveness of their instructional practices.

The Music Performance Self-Efficacy Scale (MPSES; Zelenak, 2010) was designed to measure the four sources of information (i.e., mastery experience, vicarious experience, verbal/social persuasion, and physiological state) that contribute to the development of self-efficacy beliefs (Bandura, 1986). Used as a pretest/posttest, responses from the MPSES provide formative information allowing music educators to track the development of self-efficacy beliefs among their students and offer insight into the impact of their instruction on those beliefs. The MPSES was examined in a pilot study (Zelenak, 2010). The findings were encouraging and justified a more detailed follow-up study. The current study, therefore, was a more comprehensive investigation. The purpose of this study was to examine the self-efficacy beliefs of secondary school music students through their responses to items on the MPSES and to determine whether these responses were valid and reliable representations of self-efficacy in music performance.

Bandura (1986, 1997) proposed a model of human determinism in his social cognitive theory that was driven by the reciprocal interaction of behavior, cognition, and environment. In this model, self-efficacy mediated the interaction of cognition and behavior. Bandura described self-efficacy as self-referent thought through which individuals assessed their skills and abilities to accomplish specific tasks. He also identified four sources of information that contributed to the development of self-efficacy beliefs: (a) enactive mastery experience, an individual’s past experiences of success or failure while participating in a given task; (b) vicarious experience, an individual’s observations and interpretations of others similar to one’s self engaged in the task; (c) verbal/social persuasion, judgments and opinions offered by others important to the individual regarding the behavior; and (d) physiological state, the degree and quality of arousal experienced by the individual while engaged in the activity. The information received from these four sources contributes to an individual’s self-efficacy perception for a particular task.

Scales have been developed to measure self-efficacy in a variety of subject areas. They have been constructed to measure the composite construct of self-efficacy for science (Aydin & Uzuntiryaki, 2009; Dalgety, Coll, & Jones, 2003), mathematics (Betz & Hackett, 1983), and writing (Pajares, 2007). Scales also have been created to measure the four sources of self-efficacy for academics (Usher & Pajares, 2006), mathematics (Lent, Lopez, & Bieschke, 1991; Lopez & Lent, 1992; Matsui, Matsui, & Ohnishi, 1990; Usher & Pajares, 2009), jazz music performance (Wehr-Flowers, 2007), and orchestra performance (Hendricks, 2009). The MPSES was designed to measure these same sources of self-efficacy for music performance among band, chorus, and string orchestra students at the secondary school level (i.e., Grades 6 to 12, or approximate ages 11 to 18).

Responses to self-efficacy items have been found to share common characteristics across subject areas. A comparison of results from mathematics self-efficacy scales (Lent et al., 1991; Lopez & Lent, 1992; Matsui et al., 1990; Usher & Pajares, 2009) and an academic self-efficacy scale (Usher & Pajares, 2006) indicated a common rank order among the strengths of the relationships between each source of self-efficacy and the composite construct of self-efficacy. Mastery experience consistently demonstrated the strongest relationship with self-efficacy, followed by verbal/social persuasion, physiological state, and finally, vicarious experience. Along with a common rank order, researchers have had success creating items that generated consistent responses for three of the four sources but have encountered challenges creating items that reflected vicarious experience. For example, Lent and colleagues (1991) constructed a 40-item scale with 10 items reflecting each source of self-efficacy. The internal consistency of the mastery experience items was α = .86; verbal/social persuasion, α = .74; physiological state, α = .90; and vicarious experience, α = .56. Usher and Pajares (2006) separated their vicarious experience items into peer and adult categories in an attempt to improve the internal consistency of the responses. The internal consistency of the responses to the three adult items (α = .72) was stronger than the three peer items (α = .59). I, however, believe that using only responses to adult items in an evaluation of student self-efficacy ignores the influence that peer observations may have on students’ self-efficacy.

Although other researchers (McCormick & McPherson, 2003; McPherson & McCormick, 2000, 2006) have examined the relationship between self-efficacy and music achievement, there have been few investigations of self-efficacy and music aptitude. This perspective may provide additional understandings of the interaction of self-efficacy, aptitude, and music achievement. McPherson and McCormick (2000) found self-efficacy to be the best predictor of performance achievement among students ages 9 to 18 years. From a different perspective, Fullen (1993) found students with higher levels of music aptitude demonstrated higher levels of music performance achievement. The results of these two studies raise questions regarding the relationship between self-efficacy and music aptitude and their interactive impact on music achievement.

The American Educational Research Association (AERA), American Psychological Association, and National Council on Measurement in Education (1999) have published specific standards for the evaluation of measurement instruments used in educational and psychological testing. Based on the context and intent of this study, these standards were appropriate criteria for examining the psychometric characteristics of the MPSES. The authors defined validity as “the degree to which evidence and theory support the interpretation of test scores entailed by the proposed uses of tests” (AERA et al., 1999, p. 9). Several sources were cited from which evidence of validity could be collected. These sources included test content, response processes, internal structure, and relations to other variables. Reliability was defined as “the consistency of such measurements when the testing procedure is repeated on a population of individuals or groups” (AERA et al., 1999, p. 25). The authors suggested that test developers examine the reliability of test scores based on the internal consistency of the item responses and the uniformity of scores over time.

As a pilot study, the results from Zelenak (2010) were limited. Responses from a homogenous group of participants from one school were used to investigate self-efficacy in music performance and to examine the psychometric qualities of the MPSES. Since Bandura (1986) proposed that self-efficacy was a self-perception held by all individuals regardless of personal characteristics, it was necessary to conduct an additional study to assess the item responses from the MPSES with a group of participants exhibiting greater diversity. Cronbach and Meehl (1955) advised, “A test should not be used to measure a trait until its proponent establishes that predictions made from such measures are consistent with the best available theory of the trait” (p. 262). This study was necessary, therefore, to construct accurate understandings of self-efficacy and determine whether the MPSES was an appropriate measure for use with diverse populations.

In this study, research questions consisted of the following: (1) How much influence do mastery experience, vicarious experience, verbal/social persuasion, and physiological state have on self-efficacy in music performance among secondary school music students? (2) Are there differences in the value attributed to each of the sources of self-efficacy by band, chorus, and string orchestra students? (3) Are there differences in the value attributed to the sources by middle and high school students? (4) To what extent does music aptitude predict self-efficacy in music performance? and (5) To what extent does evidence support the use of the MPSES as a valid and reliable measure of self-efficacy in music performance?

Method

Participants

One hundred twenty-four music teachers were contacted by phone or e-mail across the United States. These teachers taught in a variety of geographic, socioeconomic, and cultural contexts and were selected based upon personal knowledge of this researcher or upon recommendations by colleagues. Eleven teachers from 10 schools agreed to participate in the study. Although teachers were anxious to recruit their students as participants, many students did not return the required forms and were not allowed to participate. Only those students and teachers with completed permission and assent forms were allowed to participate in the study as required by the University of South Florida Institutional Review Board. Response rates by school were as follows: School 1, 34%; School 2, 3%; School 3, 1%; School 4, 14%; School 5, 17%; School 6, 6%; School 7, 4%; School 8, 11%; School 9, 9%; and School 10, 2%. Because of this convenience sample, the results of this study should be generalized with caution. Gall, Gall, and Borg (2007) acknowledged that participants returning parental permission forms were fundamentally different from nonparticipants. After a review of the literature, they reported that differences between participants and nonparticipants have been shown to occur in academic competency, peer popularity, physical attractiveness, and several other personal characteristics (Gall et al., 2007, p. 188).

Participants (N = 290) were middle (n =150) and high school (n = 140) music students enrolled in 10 public schools in the western (i.e., Colorado) and southeastern (i.e., Florida) regions of the United States. The participants were enrolled in band (n = 87), chorus (n = 153), and string orchestra (n = 50). Although more participants were from the Southeast (n = 227) than from the West (n = 63), the examination of regional differences was beyond the scope of this study. Male (n = 83) and female (n = 207) music students participated in the study, and results from preliminary questions indicated the participants were of varied racial backgrounds (Asian, n = 29; black, n = 29; Hispanic, n = 27; Native American, n = 2; mixed, n = 28; other, n = 8; and white, n = 167). Although these demographic descriptors indicate diversity among the participants, the proportions of these demographic descriptors may not be an exact representation of the schools’ cumulative populations. Accurate data were not available for all schools. The number of participants in the study, however, exceeded Nunnally’s (1967) recommendation for a ratio of 10 participants per item when validating a measurement scale.

Instruments

Demographic Information

Participants were asked to provide demographic information, such as birth date, gender, race, grade level, music ensemble type, and the last name of their music teacher. To maintain anonymity and security, birth dates were used to match questionnaire responses with the self-efficacy ratings provided by their teachers. In the event 2 participants had the same birth date, then the gender, race, and music teacher information was used to differentiate between participants. Grade level and ensemble type were used to classify participants into levels for analysis.

MPSES

The MPSES (Zelenak, 2010) is a 24-item instrument specifically designed to measure self-efficacy in music performance among secondary school music students participating in large ensembles (i.e., band, chorus, or orchestra; see Appendix A in online supplemental material, available at http://jrme.sagepub.com/supplemental). Items were developed to reflect Bandura’s (1986) four sources of self-efficacy. Eight items reflected mastery experience, 5 items reflected vicarious experience, 6 items reflected verbal/social persuasion, and 5 items reflected physiological state. Mastery experience and verbal/social persuasion had a greater number of items than the other sources to accommodate the generality, strength, and level of those sources as recommended by Bandura (2006). The items were presented in a random order and were not grouped by source.

The construction of the scale followed Bandura’s (2006) guidelines for measuring self-efficacy with the exception of the response format. Although Bandura recommended using an 11-point Likert-type response format (0 to 10), a 1-to-100 numerical format (1 = strongly disagree, 100 = strongly agree) was selected based on evidence of stronger internal consistency among formats of that type (Pajares, Hartley, & Valiante, 2001). Zelenak’s (2010) findings indicated that the MPSES generated responses that were internally consistent (mastery experience, α = .93; vicarious experience, α = .90; verbal/social persuasion, α = .94; and physiological state, α = .90), exhibited moderately strong correlations between pairs of sources and between each source and the composite construct of self-efficacy, and provided evidence of divergent validity by demonstrating only modest correlations with other forms of self-efficacy (e.g., writing self-efficacy and academic self-efficacy).

The MPSES used in this study was a revision of the scale used in Zelenak (2010). These changes were recommended by an expert panel of four professors and seven doctoral students in music education at a major research university. These recommendations included (a) creating separate questions for mastery experiences in solo and small-ensemble performance, (b) adding descriptive numbers to differentiate between small ensembles (2 to 10 performers) and large ensembles (11 performers or more), (c) removing a redundant item that referred to using other students as models, (d) adding a new item that asked whether the participant used someone other than a peer or professional as a model, (e) removing an item that was unrelated to the construct but was used to check a participant’s reading accuracy, and (f) rearranging the item order.

Sources of Middle School Mathematics Self-Efficacy Scale (SMSMSES)

Usher and Pajares (2009) developed the SMSMSES for use with middle school students. The same four sources of self-efficacy as the current study were measured using six items per source. Participants responded to each item by choosing a number from 1 to 100. Because of its strong psychometric qualities, Usher and Pajares recommended using the scale in other contexts. Based on this recommendation, it was appropriate to use this scale with middle and also high school students. Results from the SMSMSES were compared to results for the MPSES to establish whether the MPSES was measuring the same type of self-efficacy as the SMSMSES (i.e., mathematics self-efficacy) or a different type of self-efficacy (i.e., music performance self-efficacy).

Self-Esteem of Musical Ability (SEMA)

Schmitt (1979) constructed this 43-item scale to measure self-esteem for music. Participants responded to the items on a 4-point Likert-type scale by selecting strongly disagree, disagree, agree, or strongly agree. Using factor analysis to evaluate the responses of children 10 to 15 years of age, Schmitt identified three underlying factors: (a) self-confidence with a strong influence in music, (b) skills and abilities in music, and (c) students’ feelings of acceptance and reinforcement by parents, teachers, and friends. Since these factors and the construct of self-esteem were similar to Bandura’s (1986) concept of self-efficacy, results from this scale were compared to results from the MPSES to determine whether the MPSES was generating data that were similar to data from a related scale.

Advanced Measures of Music Audiation (AMMA)

Gordon (1989) constructed this instrument to measure music aptitude among individuals over the age of 9. He defined music aptitude as an individual’s ability to learn music. The AMMA is a listening test and consists of 30 items, with each item containing two music phrases. In the test, participants listen to two music phrases, determine whether they are the same or different, and then, if they are different, indicate whether the difference is tonal or rhythmic. Fullen (1993) found that scores from the AMMA had positive relationships with scores from a vocal performance test (7th- and 8th-grade students, r = .25; and 9th- to 12th-grade students, r = .24). Fullen suggested that music aptitude had predictive validity and could be used to predict levels of music achievement. The results from the AMMA were compared to results from the MPSES to ascertain if a relationship existed between music aptitude and self-efficacy for music performance.

Teacher Ratings

Participating teachers rated their students’ levels of self-efficacy on a 5-point Likert-type scale (1 = low, 5 = high) based on the students’ music behaviors in class. These ratings were correlated with MPSES scores to establish a relationship between the students’ self-perceptions of self-efficacy and their overt expression of the construct.

Design and Procedure

This study was descriptive in design. Data collected from the MPSES were used to construct fundamental understandings of self-efficacy in music performance and to determine whether responses from the MPSES were consistent with Bandura’s (1986) theory and the intended use of the scale. Music educators initiated the study in their ensemble classrooms by reading their students an introductory script (see Appendix B in online supplemental material, http://jrme.sagepub.com/supplemental), administering permission forms, and providing the participating students with access to computers connected to the Internet. In some classes, all participants used the computers at the same time, while in others, they used them in small groups or individually. Computers were located in the music classroom or the school media center. Every participant worked individually at a desktop computer with headphones. The questionnaires were completed during April and May of 2011. The questionnaires and audio files were hosted by an Internet-based survey company (SurveyGizmo.com). Participants completed the questionnaires in the following order: (a) demographic information, (b) MPSES, (c) SMSMSES, (d) SEMA, and (e) AMMA. The AMMA was placed last since it took the greatest amount of time to complete. Once completed, the data were downloaded as Excel files and analyzed with the software programs SAS 9.2, SPSS 19, and Mplus Version 6. Teachers returned their student ratings to the researcher as e-mail attachments or via the U.S. Post Office.

Results

Participants required 30 to 60 min to complete the battery of questionnaires. Since the questionnaires were completed during regular class periods, some participants did not complete all of the questionnaires. In addition, participants were restricted by the website software from moving freely between questionnaires. They could move from one questionnaire to the next only after completing the current questionnaire. All participants completed the MPSES (n = 290); most completed the SMSMSES (n = 289) and the SEMA (n = 286), and a large number completed the AMMA (n = 222). The relationship between self-efficacy (MPSES) and music aptitude (AMMA) was established using scores from only those participants who completed both questionnaires.

Data from the MPSES were examined to determine whether they met the assumptions for multivariate analysis. Evidence of independence was based on the premise that the MPSES was a self-report measure, and it was likely that each participant had a unique history of music experiences. Further evidence was provided through bivariate scatterplots in which all possible pairs of the sources of self-efficacy were plotted on x- and y-axes. Only the plot of mastery experience by verbal/social persuasion had elliptical characteristics, indicating some correlation between those sources. Indicators of normality, however, were contradictory. Univariate skewness and kurtosis values (γ₁ = −0.70 and γ₂ = 0.12) suggested a normal distribution, while multivariate values suggested non-normality: mulitvariate skewness, b1p = 5.33, χ²(20, N = 290) = 261.52, p < .05, and b2p = 32.79, 95% CI [10.28, 10.80]. Stevens (2009) advised that the F statistic was robust to violations of normality in ANOVA and MANOVA procedures and that the primary consequence of non-normality was a reduction in power. Problems also arose in establishing homogeneity of the covariance matrices. Stevens, however, found that the MANOVA was robust to violations of the homogeneity-of-variance and homogeneity-of-covariance assumptions as long as sample sizes were similar, the data were interval level, and the sample size was large (N ≥ 200). In this study, the grade-level group sizes were approximately equal (middle school, n = 150; high school, n = 140), the data were interval level, and the sample size was large (N = 290). The only exception was that the ensemble sizes were unequal (band, n = 87; chorus, n = 153; and string orchestra, n = 50). Due to these irregularities, Pillai’s trace was used as the outcome statistic from the MANOVA rather than Wilks’s lambda based on its tendencies as a more conservative indicator.

Sources of Self-Efficacy

Confirmatory factor analysis (CFA) was used to estimate the loadings (i.e., degree of influence) of the sources of self-efficacy on the composite construct of self-efficacy. In CFA, Bandura’s (1986) theoretical framework determined the structure of the factor model. Scale items were then used to reflect the four sources of self-efficacy. Results indicated that mastery experience had the strongest loading (StdYX = .98), followed by verbal/social persuasion (StdYX = .91), physiological state (StdYX = .83), and vicarious experience (StdYX = .75) (see Figure S1 in online supplemental material, available at http://jrme.sagepub.com/supplemental). The interpretation of the StdYX value is similar to the interpretation of a standardized regression coefficient. The StdYX values indicated that mastery experience had a greater similarity to the composite construct than the other sources and that a one-unit change in mastery experience resulted in approximately a one-unit change in self-efficacy.

MANOVA was used to determine whether there were differences in the sources of self-efficacy among participants by ensemble, by grade level, and by the interaction of ensemble and grade level. MANOVA is an extension of the univariate ANOVA procedure and is used to compare group means on multiple dependent variables. MANOVA has two primary advantages over using multiple ANOVA procedures: (a) greater effectiveness in controlling Type I error and (b) the ability to incorporate correlations among the variables into the analytical model. The scores from the items reflecting each source of self-efficacy were combined and analyzed as separate dependent variables. Items 1, 4, 6, 8, 10, 12, 14, and 16 reflected mastery experience; Items 2, 5, 11, 18, and 20 reflected vicarious experience; Items 3, 7, 9, 13, 21, and 22 reflected verbal/social persuasion; and Items 15, 17, 19, 23, and 24 reflected physiological state. Based on the scores from these dependent variables, MANOVA results indicated no significant difference among ensembles, F(8, 564) = 1.39, p = .20, V = 0.04; no difference between grade levels, F(4, 281) = 1.76, p = .14, V = 0.02; and no difference in the interaction of ensemble and grade level, F(8, 564) = 1.89, p = .06, V = 0.05. Participants provided similar responses to the items on the MPSES regardless of ensemble, grade level, or interaction of ensemble and grade level.

Linear regression was used to calculate the percentage of self-efficacy in music performance that could be predicted given an individual’s music aptitude. Linear regression is the simplest form of regression analysis, in which a single predictor variable is used to calculate a criterion variable. Responses from Gordon’s (1989) AMMA were combined and used as the predictor variable, and responses from the MPSES were combined and used as the criterion variable. To allow for comparisons with other regression coefficients, the participants’ combined scores were standardized (i.e., transformed into z scores) prior to constructing the equation.

AMMA scores in this study were comparable to those reported in other studies. The mean score for high school students in this study (M = 50.99) was similar to Gordon’s (1989) reported mean (M = 50.60), and the mean score for middle school students in this study (M = 43.78) was somewhat lower than the mean reported by Gordon (2004; M = 46.10). Regression results indicated that music aptitude scores predicted a modest increase in self-efficacy scores, β = .16 (SE = .07). Students with higher levels of music aptitude tended to have slightly higher levels of self-efficacy (r = .02). From the regression equation, the proportion of predictable variance (r²) indicated that approximately 3% of the variance in self-efficacy could be predicted from music aptitude. Although this relationship was significant, F(1, 220) = 6.12, p = .01, 95% CI [.03, .29], the proportion of predictable variance was very small.

Validation of the MPSES

Criteria established by AERA et al. (1999) were used to evaluate the MPSES. Those criteria included establishing evidence of test content, response process, internal structure, and relationships to other variables. To evaluate the content of the MPSES, a panel of four professors and seven doctoral students in music education at a comprehensive research university examined the scale items. After reviewing the revised MPSES items, this panel of experts believed that the items on the MPSES accurately reflected Bandura’s (1986) construct of self-efficacy as applied to music performance in the secondary school classroom. To establish evidence of the response process, eight music teachers evaluated the self-efficacy of their participant students based on personal observations of their students’ in-class music behaviors. Teachers rated their students’ (n = 175) self-efficacy for music performance on a 5-point Likert-type scale (1 = very low, 5 = very high). Although there was great variety in the ability of teachers to make this determination (individual teacher’s correlations ranged from r = –.09 to r = .66), combined results suggested a moderately positive relationship between teacher ratings and student scores (r = .42, p < .001). This relationship was similar to the relationship found in Zelenak (2010; r = .44) and provided evidence that the self-report response process of the MPSES generated data that were moderately consistent with the participant’s observed behavior.

Results from the CFA provided evidence of good fit between responses generated by the MPSES and Bandura’s (1986) theory. A second-order CFA was used to evaluate the fit. This second-order CFA consisted of a first order with relationships established between the scale items and the sources of self-efficacy and a second order with relationships established between the sources of self-efficacy and the composite construct (see Figure S1 in the online supplemental material, available at http://jrme.sagepub.com/supplemental). To identify the model, loadings were set to 1.0 between one item and each source of self-efficacy and also between the source of mastery experience and the composite self-efficacy. Maximum likelihood estimation was used to estimate the other loadings, errors, and variances. Initial CFA results, χ²(248, N = 290) = 568.09, p = .001, comparative fit index (CFI) = .84, standardized root mean square residual (SRMR) = .06, and root mean square error of approximation (RMSEA) = .07, were contradictory, indicating acceptable fit by the SRMR but not by the RMSEA or other indices. Hu and Bentler (1998) recommended the benchmarks of SRMR < .08 and RMSEA < .06 as indicators of good fit. After making minor modifications (i.e., correlating the errors of three items), the CFA produced results indicating good fit between responses on the MPSES and Bandura’s (1986) theoretical model, χ²(245, N = 290) = 501.62, p = .001, CFI = .87, SRMR = .06, RMSEA = .06.

Evidence of relationships to other variables was established from discriminant and convergent perspectives. From a discriminant perspective, a modest relationship was found between scores on the MPSES and scores on the SMSMSES (r = .29, p < .001). The strength of this relationship suggested some commonality in the construct measured by each instrument, but fundamentally, the MPSES and SMSMSES were measuring different constructs. From a convergent perspective, MPSES scores exhibited a moderately strong relationship with scores on the SEMA (r = .67, p < .001). The strength of this relationship provided evidence that results from the MPSES were comparable to results from the SEMA. In addition, the correlation between teacher ratings and student self-efficacy scores (r = .42, p < .001) provided evidence of multimethod convergence. Teacher ratings were based on observable behaviors, while student scores were generated from a self-report questionnaire.

AERA et al. (1999) also recommended examining the reliability of educational measurement instruments through the internal consistency of responses and the stability of scores over a period of time. The internal consistency of the combined responses on the MPSES (α = .88) exceeded the common benchmark of α =.80 for general research purposes (Henson, 2001; Nunnally, 1967). Among the individual sources of self-efficacy, responses to items reflecting verbal/social persuasion were the most consistent (α = .77), followed by mastery experience (α = .74) and physiological state (α = .67). Responses to items reflecting vicarious experiences were the least consistent (α = .59). The low level of internal consistency for items reflecting vicarious experiences mirrored problems reported in other self-efficacy studies (Lent et al., 1991; Usher & Pajares, 2006). To evaluate the stability of scores over time, a small group of students (n = 14) completed the MPSES 3 weeks after the first administration. With this small group, scores were stable over time (r = .87). Unfortunately, this study was conducted at the end of the academic school year, and it was not possible to obtain a larger number of participants for the test-retest assessment.

Discussion

The purpose of this study was to examine the self-efficacy beliefs of secondary school music students through their responses to items on the MPSES and to determine whether these responses were valid and reliable representations of self-efficacy in music performance. The first research question addressed the degree of influence each source of self-efficacy had on the composite construct. Results indicated that mastery experience had the greatest degree of influence, followed by verbal/social persuasion, physiological state, and vicarious experience. This rank order among the sources was consistent with the findings in other subject areas (Lent et al., 1991; Lopez & Lent, 1992; Matsui et al., 1990; Usher & Pajares, 2006, 2009). In addition, the conclusion that mastery experience exerted the greatest influence on the development of self-efficacy was consistent with Bandura’s (1986) theory.

The results from the first research question offer insights for educators and researchers. For educators, the powerful influence of mastery experiences on student self-efficacy must not be overlooked. One application in the classroom would be to refrain from requiring students to perform music beyond their ability level. This practice may have a negative influence on student self-efficacy, resulting in poor performances and a lack of interest in music. Educators should carefully choose repertoire that matches their students’ ability levels. This finding supports the practice of organizing ensembles based on ability and offering differentiated instruction opportunities for all students.

Researchers may find these results valuable in several ways. From one perspective, the influence of these four sources may not be limited to self-efficacy. These same sources may influence the ways in which music students view themselves, their place in society, their reasons for performing music, and the effort they put forth to reach higher levels of achievement. This perspective may interest researchers of self-identity, goal orientation, and motivation. From another perspective, the shared rank order of the sources in music and other subject areas allows music education researchers the opportunity to examine the findings of self-efficacy from other subjects and the influence of their integration into music classroom. For example, Usher and Pajares (2009) identified the importance of incremental opportunities to experience success in mathematics, while Usher and Pajares (2006) recommended using positive verbal reinforcement that was genuine and authentic in academics, and Dalgety et al. (2003) proposed using a self-efficacy scale to identify specific topics that students found difficult in science. Literature reviews or replication studies on issues particularly relevant to music education also may be driven by this finding.

The second and third research questions called for comparisons of the responses to items on the MPSES among participants in different ensembles (i.e., band, chorus, and string orchestra) and between participants in different grade levels (i.e., middle and high school). Results indicated that participants provided similar responses to MPSES items regardless of ensemble, grade level, or the interaction of ensemble and grade level. In other words, all participants reported similar self-perceptions of their music experiences. Based on this finding, comparable instructional activities may be used in the music classroom by band, chorus, and string orchestra educators at the middle or high school grade level with similar impact on student self-efficacy. These activities may include the use of performance opportunities to promote positive mastery experiences (i.e., performing for receptive audiences with repertoire that matches students’ ability levels), using peer models to foster vicarious experiences, encouraging positive feedback from friends and family members, and teaching students to control stress and anxiety during performances to reach more relaxed psychological states. In addition, the MPSES can be used by educators to monitor the impact of their instructional activities in the ensemble classroom and can provide the educator with quantitative feedback regarding the effectiveness of their instructional choices. Scores from the MPSES can be used to identify strengths and weaknesses of the beliefs related to each source of self-efficacy.

The fourth research question prompted an examination of the relationship between music aptitude and self-efficacy in music performance. Results indicated a modest positive relationship between music aptitude and music self-efficacy. This finding does not imply that music aptitude causes higher levels of self-efficacy but, rather, that there is a relationship between the two constructs. Individuals with higher levels of music aptitude tended to have slightly higher levels of self-efficacy. On one hand, this finding is important because it quantifies the relationship between the constructs of self-efficacy and music aptitude. On the other hand, this finding provides a theoretical connection between the literature documenting Gordon’s (1993) music learning theory and Bandura’s (1986) social cognitive theory. Since there are substantial bodies of literature focused on these theories, the importance of this connection cannot be overstated. Psychological constructs are not isolated but interactive.

The cumulative findings of this study provide evidence to support the use of the MPSES as a valid and reliable instrument for assessing self-efficacy in music performance. Several forms of evidence were established based on the recommendations of AERA et al. (1999). Content validity was confirmed by experts in the field. The effectiveness of the response process was established by examining the relationship between scores from the MPSES with teacher ratings of the participants. It is important to note, however, that some teachers were better at determining their students’ self-efficacy levels than others. Educating teachers about the overt expressions of self-efficacy may be another way of incorporating the findings of this study into the classroom. An additional form of evidence came from an investigation into the internal structure of the MPSES. This investigation determined that data generated from the MPSES had good fit with Bandura’s (1986) theoretical model. As further evidence of the internal structure, the factor loadings of the four sources of self-efficacy onto the composite construct were consistent with theory and findings from studies in other subject areas.

Finally, evidence was established that defined the relationship of scores from the MPSES with those of other variables. Scores from the MPSES exhibited a relatively weak positive relationship with scores from Usher and Pajares’s (2009) SMSMSES, indicating the two scales measured constructs that were somewhat different. Conversely, scores from the MPSES demonstrated a moderately strong positive relationship with scores from Schmitt’s (1979) SEMA, indicating commonality between the two constructs. On the basis of these correlations, items on the MPSES elicited responses that demonstrated evidence of discriminant and convergent validity.

Along with establishing validity for the intended use of the MPSES, the reliability of responses from the MPSES also was examined in this study. The item responses exhibited a high degree of internal consistency even though the items reflected self-efficacy with different levels of generality and strength. Participants provided similar responses to items associated with each source of self-efficacy. Further investigation of vicarious experience, however, is warranted and may contribute information useful in developing a subscale with greater reliability. From another perspective, MPSES items generated consistent responses over a 3-week time period. Although the number of individuals participating in the test-retest procedure was small (n = 14), this finding suggests that items reflect a construct that is stable over time. Further testing with a larger group is recommended. This collection of findings, nonetheless, provided substantial evidence that the MPSES met the criteria for validity and reliability as recommended by AERA et al. (1999).

In this study, I attempted to make a contribution to filling the gap in the understanding of self-efficacy in music performance. From this vantage point, additional research might include the investigation of the self-efficacy beliefs of music students in different age groups, such as undergraduates or elementary-level students. These investigations might prompt further questions about the ways in which music self-efficacy develops with age. The relationships between self-efficacy, the sources of self-efficacy, and music achievement also need to be investigated. Since the MPSES was found to generate valid and reliable data, this measurement instrument could be used with a high degree of confidence in further quantitative investigations of self-efficacy along with providing supportive evidence for examinations of other psychological constructs, such as self-identity, motivation, or self-regulation. Hopefully, authors of future studies will develop new understandings of the ways in which individuals respond to, create, and perform music. In this way, the music education community can continue to pursue its agenda of assisting students attain higher levels of achievement.

Footnotes

Author’s Note

This article is based on the author’s dissertation, “Self-Efficacy in Music Performance: Measuring the Sources Among Secondary School Music Students,” completed at the University of South Florida in 2011.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Supplemental Material

The online supplemental material is available at .

Author Biography

Michael S. Zelenak is assistant professor of music education at Alabama State University. His research interests include the self-perceptions associated with music performance and the measurement of music learning among K–12 students.

References

American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (1999). Standards for educational and psychological testing. Washington, DC: American Educational Research Association.

Austin

J. R.

Vispoel

W. P.

(1998). How American adolescents interpret success and failure in classroom music: Relationships among attributional beliefs, self-concept, and achievement. Psychology of Music, 26, 26–45. doi:10.1177/0305735698261004

Aydin

Y. C.

Uzuntiryaki

(2009). Development and psychometric evaluation of the high school chemistry self-efficacy scale. Educational and Psychological Measurement, 69, 868–880. doi:10.1177/0013164409332213

Bandura

(1986). Social foundations of thought and action. Englewood Cliffs, NJ: Prentice Hall.

Bandura

(1997). Self-efficacy: The exercise of control. New York, NY: Freeman.

Bandura

(2006). Self-efficacy beliefs of adolescents. Information Age.

Betz

N. E.

Hackett

(1983). The relationship of mathematics self-efficacy expectations to the selection of science-based college majors. Journal of Vocational Behavior, 23, 329–345.

Cronbach

L. J.

Meehl

P. E.

(1955). Construct validity in psychological tests. Psychological Bulletin, 52, 281–302.

Dalgety

Coll

R. K.

Jones

(2003). Development of chemistry attitudes and experiences questionnaire. Journal of Research in Science Teaching, 40, 649–668. doi:10.1002/tea.10103

10.

Fullen

D. L.

(1993). An investigation of the validity of the Advanced Measures of Music Audiation with junior high and senior high school students (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (AAT 9316479)

11.

Gall

M. D.

Gall

J. P.

Borg

W. R.

(2007). Educational research: An introduction. New York, NY: Pearson Education.

12.

Gordon

E. E.

(1989) Advanced measures of music audiation. Chicago, IL: GIA.

13.

Gordon

E. E.

(1993). Learning sequences in music: Skill, content, patterns. A music learning theory. Chicago, IL: GIA.

14.

Gordon

E. E.

(2004). Continuing studies in music aptitudes. Chicago, IL: GIA Publications, Inc.

15.

Greenberg

(1970). Musical achievement and the self-concept. Journal of Research in Music Education, 18, 57–64.

16.

Hendricks

K. S.

(2009). Relationships between the sources of self-efficacy and changes in competence perceptions of music students during an all-state orchestra event (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (AAT 3362920)

17.

Henson

R. K.

(2001). Understanding internal consistency reliability estimates: A conceptual primer on coefficient alpha. Measurement and Evaluation in Counseling and Development, 34, 177–189.

18.

Bentler

P. M.

(1998). Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods, 3, 424–453. doi:10.1037/1082-989X.3.4.424

19.

Lent

R. W.

Lopez

F. G.

Bieschke

K. J.

(1991). Mathematics self-efficacy: Sources and relation to science-based career choice. Journal of Counseling Psychology, 38, 424–430.

20.

Lopez

F. G.

Lent

R. W.

(1992). Sources of mathematics self-efficacy in high school students. Career Development Quarterly, 41, 3–9.

21.

Lum

(2008). Home musical environment of children in Singapore: On globalization, technology, and media. Journal of Research in Music Education, 56, 101–117. doi:10.1177/0022429408317517

22.

Martin

(2012). The musical self-efficacy beliefs of middle school band students: An investigation of sources, meanings, and relationships with attributions for success and failure. Bulletin of the Council for Research in Music Education, 191, 45–60.

23.

Matsui

Ohnishi

(1990). Mechanisms underlying math self-efficacy learning of college students. Journal of Vocational Behavior, 37, 225–238.

24.

McCormick

McPherson

(2003). The role of self-efficacy in a musical performance examination. Psychology of Music, 31, 37–51. doi:10.1177/0305735603031001322

25.

McPherson

G. E.

McCormick

(2000). The contribution of motivational factors to instrumental performance in a music examination. Psychology of Music, 31, 37–51. doi:10.1177/1321103X0001500105

26.

McPherson

G. E.

McCormick

(2006). Self-efficacy and music performance. Psychology of Music, 34, 322–336. doi:10.1177/0305735606064841

27.

McPherson

G. E.

Zimmerman

B. J.

(2002). Self-regulation of musical learning: A social cognitive perspective. In Colwell

Richardson

(Eds.), The new handbook of research in music teaching and learning (pp. 327–347). New York, NY: Oxford University Press.

28.

Miksza

(2011). The development of a measure of self-regulated practice behavior for beginning and intermediate instrumental music students. Journal of Research in Music Education, 59, 321–338. doi:10.1177/0022429411414717

29.

Nielsen

S. W.

(2004). Strategies and self-efficacy beliefs in instrumental and vocal individual practice: A study of students in higher music education. Psychology of Music, 32, 418–431. doi:10.1177/0305735604046009

30.

Nunnally

J. C.

(1967). Psychometric theory. New York, NY: McGraw-Hill.

31.

Pajares

(2007). Empirical properties of a scale to assess writing self-efficacy in school contexts. Measurement and Evaluation in Counseling and Development, 39, 239–249.

32.

Pajares

Hartley

Valiante

(2001). Response format in writing self-efficacy assessment: Greater discrimination increases prediction. Measurement and Evaluation in Counseling and Development, 33, 214–221.

33.

Randles

(2010). The relationship of compositional experiences of high school instrumentalists to music self-concept. Bulletin of the Council for Research in Music Education, 184, 7–18.

34.

Ritchie

Williamon

(2007). Measuring self-efficacy in music. In International Symposium on Performance Science (pp. 307–312). Utrecht, Netherlands: European Association of Conservatoires (AEC).

35.

Schmitt

M. C. J.

(1979). Development and validation of a measure of self-esteem of musical ability (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (AAT 8009164)

36.

Stevens

J. P.

(2009). Applied multivariate statistics for the social sciences. New York, NY: Routledge/Taylor & Francis Group.

37.

Svengalis

J. N.

(1978). Music attitude and the preadolescent male (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (AAT 7902953)

38.

Usher

E. L.

Pajares

(2006). Sources of academic and self-regulatory efficacy beliefs of entering middle school students. Contemporary Educational Psychology, 31, 125–141. doi:10.1016/j.cedpsych.2005.03.002

39.

Usher

E. L.

Pajares

(2009). Sources of self-efficacy in mathematics: A validation study. Contemporary Educational Psychology, 34, 89–101. doi:10.1016/j.cedpsych.2008.09.002

40.

Wehr-Flowers

E. L.

(2007). An exploratory model of jazz self-efficacy and gender (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (AAT 3281418)

41.

Zelenak

M. S.

(2010). Development and validation of the Music Performance Self-Efficacy Scale. Music Education Research International, 4. Retrieved from http://cmer.arts.usf.edu/content/articlefiles/3122-MERI04pp31-43.pdf

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB

0.02 MB

0.04 MB