Abstract
The purpose of this study was to examine the effect of conductor expressivity on the evaluation of rehearsal instruction. We video recorded two conductors in rehearsal with a university band. We extracted a 3.5-min excerpt of their rehearsals that featured alternation between conductor talk and expressive conducting. For one of the conductors (the “experimental” conductor), we also re-recorded the same rehearsal excerpt with the conductor using unexpressive gestures. In postproduction, we created two versions of the experimental conductor’s rehearsal—one unaltered and the other altered to show unexpressive conducting gestures. The footage of the experimental conductor’s verbal instruction was identical in both versions. Collegiate musicians (N = 134) viewed the rehearsal excerpts of both the experimental conductor (either expressive or unexpressive) and the control conductor and evaluated both conductors on eight criteria related to instructional effectiveness. Significant differences were found between conditions in the composite evaluations of the experimental conductor. Further inspection revealed significantly higher evaluations favoring the expressive condition on seven of the eight criteria. Notable among these results is the influence of expressive gesture on the perception of distinctly verbal conductor behaviors. We suggest this may indicate the inseparability of gestural and verbal aspects of rehearsal instruction.
In school music programs, ensemble conductors serve an iconic role as instructional leaders. Teachers in no other subject matter would seem to be so explicitly reliant on nonverbal communication as a primary means of conveying academic expectations for students during both in-class instruction (i.e., rehearsals) and public demonstrations of summative achievement (performance). The importance of conducting as a discrete program of study is reflected in its long-standing and seemingly universal inclusion as a requirement for degree programs in music education. In fact, Labuta (1965) deemed conducting to be the most important pedagogical competency for music education majors to learn. Professionals have long concurred that conductors’ expressive nonverbal skills are important (Julian, 1989; Krudop, 2003) and that particular components thereof are definable and learnable (Ostling, 1976). Other research findings (Manfredo, 2008; Romines, 2003) indicate that instruction in gesture often takes precedence over rehearsal skills within undergraduate conducting courses.
Expert teachers employ a more varied repertoire of nonverbal instructional skills, including expressive gesture (Byo & Austin, 1994; Johnson, Darrow, & Eason, 2008), and they have a more fluent command of conducting behaviors toward specific ends in music teaching compared with less experienced teachers (Bergee, 2005). Insistence on expressive performance is also a comparatively more frequent element of experts’ verbal instruction (Bergee, 2005; Goolsby, 1999). As such, expression itself appears to be both a means and an end in effective music teaching, and the deliberate use of gesture seems crucial in this regard. Some researchers have explored the direct influence of gesture—that is, young musicians’ own use of gesture or their own experience with conducting—on their subsequent performance success, with positive effects found among young singers (Liao, 2008), beginning band students (Cofer, 1998; Kelly, 1997), and high school string orchestra members (Thompson, 2012).
The importance of expressive conducting specifically and high-affect instruction more generally is reflected in their inclusion as aspects of music teacher evaluation scales with established validity and reliability (Bergee, 1992; Hamann & Baker, 2004). Nonverbal components of high-magnitude music teaching include sustained eye contact, varied and animated facial expressions and gestures, movement on the podium, and a brisk rehearsal pace. Such elements have been associated with effective teaching in the general education research literature (Brophy, 1987; McCroskey, Richmond, Sallinen, Fayer, & Barraclough, 1995) and within music education scholarship as well (Butler, 2001; Hamann, Baker, McAllister, & Bauer, 2000; Hendel, 1995; C. K. Madsen & Geringer, 1989; K. Madsen, 2003; Saunders & Worthington, 1990; VanWeelden, 2002; Yarbrough, 1975). Yarbrough (1975, p. 138) offered operational definitions of six elements of high- and low-magnitude teacher behavior, including conducting-related behaviors such as gestures (“Uses arms and hands to aid in musical phrasing. Great variety of movement. Varies size of conducting patterns to indicate phrases, dynamics, and the like” vs. “Strict conducting pattern, never varying”) and facial expressions (“Face reflects sharp contrasts between approval/disapproval . . . ,” as opposed to “Neutral mask. No frowns. No smiles.”). Very similar definitions were given by subsequent researchers in their investigations of expressive versus nonexpressive conducting (e.g., House, 1998; Laib, 1993; Mayne, 1992; Sidoti, 1990) and employed in the current study as well.
Results on the direct effects of various conducting behaviors are mixed. Grechesky (1985) found positive associations between several elements of expressive conducting (e.g., body movement, approving facial expressions, left hand usage) and ensemble performance among high school bands. Also, in some experimental studies, expressive conducting has yielded higher performance ratings versus nonexpressive conducting among individual high school (House, 1998) and collegiate (Sidoti, 1990) instrumentalists and in band performance (Laib, 1993). Rather consistent findings have been reported regarding preference for expressive conducting conditions (or components thereof) among ensemble members of varying genres and experience levels and observers of ensemble performance (Harden, 2000; House, 1998; Laib, 1993; Peddell, 2008; Price & Mann, 2011; Price & Winter, 1991; Silvey & Koerner, 2016; Whitaker, 2011), with particular elements of nonverbal technique (e.g., right arm movement, left arm movement, eye contact, posture) being of varying levels of importance (Johnson, Fredrickson, Achey, & Gentry, 2003; VanWeelden, 2002). However, researchers in some of these same studies found no effect of expressive conducting on secondary-level bands’ performance quality (Price & Winter, 1991; Silvey & Koerner, 2016). In a series of studies in which audio recordings of ensemble performance and video recordings of those ensembles’ conductors were evaluated separately, no relationship was found between conductor expressivity and ensemble expressivity (Price & Chang, 2001, 2005) or overall conductor quality and overall ensemble performance (Price, 2006).
Further investigations have revealed a more nuanced and complicated set of relationships among expressive gesture, expressive performance, and the perception of visual and aural information among observers. Physical movement itself appears to convey a sense of musical expression to undergraduate musician observers (Davidson, 1993). In a study of music majors’ evaluations of a pianist’s simulated performance, ratings of dynamics, phrasing, and overall musical effect were highest in full-body movement conditions compared to head-and-facial movement and no-movement conditions (Juchniewicz, 2008). Similar results have been found in evaluations of ensemble expressivity in band performance as evaluated by music majors (Morrison, Price, Geiger, & Cornacchio, 2009) and nonmajors (Price, Mann, & Morrison, 2016), choral settings among both musically experienced and less experienced observers (Morrison & Selvey, 2014), and evaluations of classical music competitions (Tsay, 2013) and professional orchestras (Tsay, 2014) by both professional musicians and musically untrained persons. The presence or absence of conductor facial expression appears to affect perceived ensemble expressivity (Silvey, 2013), as does the angle from which one views an ensemble conductor (Napoles, 2013). Ratings of other specific aspects of ensemble performance (dynamics and articulation) have also been found to be highest when conductor gesture matches the intended effect (Morrison, Price, Smedley, & Meals, 2014). Expressive teaching (i.e., demonstrated teacher intensity behaviors) influences the perception of ensemble performance quality (Kaiser, 1998). Together, these findings highlight the contribution of many different components of visual information in the evaluation of musical events.
Less is known about the relationship between gestural and verbal elements of instruction in the perception of music teaching effectiveness. Variables extraneous to the teacher or conductor, such as student on- or off-task behavior (K. Madsen, 2003) and ensemble performance quality (Bender & Hancock, 2010; Silvey, 2011), do appear to influence the evaluation of musical instruction. Exploring aspects of expressive conducting from a rehearsal perspective (rather than only by virtue of the performance of the ensemble) would add a meaningful and contextual foundation from which to gather additional information about expressivity for researchers, conducting pedagogues, and practitioners. Thus, we wanted to test whether conductor expressivity would influence musicians’ evaluations of teaching effectiveness in a rehearsal setting while attempting to focus observers’ attention specifically on verbal (i.e., nongestural) aspects of the teacher’s behavior in rehearsal. The purpose of this study was to examine the effect of conductor expressivity on the evaluation of rehearsal instruction. Would college musicians rate the teaching effectiveness of a conductor differently based on how expressively they conducted?
Method
Synopsis
We video recorded excerpts of two conductors in rehearsal with a university band. For the second conductor, we re-recorded the same rehearsal excerpt with the conductor using unexpressive gestures. In postproduction, we created two versions of this conductor’s rehearsal (one unaltered and expressive, the other altered and unexpressive). We presented the video recordings of both conductors to collegiate musicians (with half viewing the unaltered version and the other half viewing the altered version of the second conductor) via an online survey platform and asked them to evaluate verbal (i.e., nongestural) aspects of their teaching. Prior to participant (i.e., evaluator) recruitment, we used G*Power 3.1.9.2 software (Faul, Erdfelder, Lang, & Buchner, 2007) to determine the necessary sample size for ensuring statistical power of .80 (Cohen, 1988). Given our plan to use a between-subjects ANOVA and a significance level of α = .05 and anticipating a medium effect size of .25 (Cohen’s f), we calculated a minimum sample size of 128.
Participants
Evaluators
Participants (N = 134) in this study were undergraduate (n = 125) and graduate (n = 7) college musicians who were currently performing in auditioned concert bands at 1 of 12 university schools of music located throughout the United States. The respective faculty conductors of each of these ensembles announced the study during rehearsal and again via an email to all ensemble members. Other participants were recruited via email from among other instrumental music education and music performance majors at these same institutions. There were 77 male and 53 female participants as well as 2 who indicated “other or prefer not to answer” who reported their age (M = 20.7 years, SD = 5.8; all but 3 participants were under 34 years old), performing emphasis (woodwind n = 68; brass n = 40; percussion n = 11; string n = 4; voice n = 4, piano/keyboard n = 5), degree program (music education n = 38; music performance n = 10; other music degree n = 4; non–music major n = 80), and years of performing in a conducted ensemble (M = 9.4, SD = 4.5). Two participants declined to provide demographic information.
Study conductors
Two female graduate students (ages 42 and 31) were chosen to serve as rehearsal conductors. At the time of the study, both conductors were enrolled in a graduate wind conducting program at a large Rocky Mountain university school of music. Prior to their selection and commitment to the project, two university conducting faculty members (who conducted bands at the second author’s institution and did not know the study conductors) viewed videotaped rehearsal footage of both conductors. We asked these experts to verify that both conductors demonstrated expert teaching and conducting behaviors, as indicated in previous research literature (Duke & Simmons, 2006; Goolsby, 1996; Madsen & Geringer, 1989; Yarbrough, 1975). A printed list of these characteristics was provided to both faculty members. Examples of verbal rehearsal behaviors included quick alternation between conductor talk and ensemble performance, specific feedback, and rehearsal of musical skills such as intonation, balance, and blend. Expressive conducting traits included left hand independence (cf. Chapman, 2008), fluid gestures that varied according to the style and dynamic shape of the music, specific eye contact, and varied facial expressions (Byo & Austin, 1994). Both experts agreed that the conductors shown in the videos demonstrated these traits.
Ensemble performers
The symphonic band used in this study was the second auditioned ensemble (of three) at a large Rocky Mountain university school of music. Members (N = 65) ranged in age from 18 to 25; all but two were undergraduates. None of the ensemble performers participated later as evaluators.
Repertoire Selection and Conductor Preparation
One conductor was assigned to rehearse Jubilee Overture by Philip Sparke, with the other given Valley Montage and Celebration by Richard Saucedo. These excerpts were identified by their publisher, Hal Leonard, as either grade 4 or grade 5 (out of 6), respectively. Similar to the rationale provided by Silvey and Baumgartner (2016), these selections were chosen because they would not be played flawlessly during brief ensemble rehearsals, allowing each conductor to respond to musical issues that typically arise in ensemble rehearsals (e.g., rhythmic precision, balance, blend, intonation, phrasing). Excerpts were approximately 2 min in duration and featured contrasting tempi and opportunities for phrase shaping and dynamic contrasts, characteristics that are consistent with recent conductor expressivity methodologies (Morrison & Selvey, 2014; Price et al., 2016; Silvey & Fisher, 2015).
Conductors were given scores to their pieces and told they would be leading an approximately 15-min rehearsal of their assigned excerpt with the symphonic band, an ensemble that was familiar to both conductors due to their previous conducting assignments with the group. To aid in their score study and rehearsal planning processes, each conductor was also given a model recording of their assigned excerpt. Conductors were informed that they would have 1 week to study their scores before leading their rehearsals and that they should focus on conducting their excerpts expressively. As a means of addressing ecological validity, we told the conductors that they should address whatever musical issues they believed needed remediation based on their personal score study and the performance of the ensemble.
Video Recordings
Digital video recordings of conductors’ rehearsals were made during the first half-hour of a regularly scheduled symphonic band meeting. Two professional-quality video cameras (a Sony FS100 NXCam and a Canon 5D Mark II) were stationed in the rehearsal hall, approximately 25 ft. from the podium in the back of the rehearsal space, behind the percussion section. The cameras recorded left and right side–angle views of the conductor. Both cameras were trained toward the conductor; only a partial view of players in the front row of the ensemble was captured in the viewfinder. The conductor who was assigned Jubilee Overture (hereafter referred to as the “experimental conductor”) rehearsed first, followed by the conductor who was assigned Valley Montage and Celebration (hereafter referred to as the “control conductor”). Other rehearsal selections unrelated to the current study took place following our two conductors’ rehearsals, which were led by other conductors. A third camera, also positioned in the back of the rehearsal hall and out of the view of the three main cameras, recorded the rehearsal of the experimental conductor only. A professional-quality audio microphone (from a Zoom Q8 video recorder) was placed near the podium to better capture the conductors’ verbalizations and the ensemble’s performance.
Immediately following the first rehearsal selection, the primary author met with the experimental conductor in an adjacent classroom to view the video footage from her rehearsal as recorded on the third video camera. Then, during the last 15 min of the allotted rehearsal time for the day, the experimental conductor returned to the podium. She simulated her previous rehearsal of Jubilee Overture, this time conducting in a strictly unexpressive manner, with a consistently neutral facial expression, minimal eye contact (i.e., looking at the conductor score or looking generally across the ensemble and not to a specific section), and with conducting gestures that only followed the beat pattern and did not vary according to the dynamic shapes of musical phrases (Byo & Austin, 1994; House, 1998; Sidoti, 1990; Yarbrough, 1975). By manipulating these elements as such in this simulation, we sought to feature (inasmuch as possible) technically correct but manifestly nonexpressive conducting—gestures that followed pattern conventions and were clear to follow but provided no information regarding the feelingful nature of the music. During this time, the audio recordings of her earlier rehearsal were played for the ensemble over mounted stereo loudspeakers, and the ensemble members were instructed to simulate playing along with her conducting and follow her stopping and restarting of the ensemble as heard on the audio recording. The conductor’s attire, the ensemble members’ attire and seating positions, and the camera angles were all unchanged from the earlier portion of the rehearsal.
Audiovisual Stimulus Creation
A professional videographer with over 10 years of audio and video engineering experience assisted with the video editing process. For both the experimental and the control conductors, one 3.5-min excerpt was chosen from the footage of each of their respective rehearsals. These excerpts encompassed alternating segments of conducting and ensemble playing and verbal instruction from the podium, both of varying durations. In postproduction, we edited the videos in a manner that featured the conductor from both camera angles throughout the rehearsal segment. Then, for the experimental conductor only, we created an additional, alternative version of her 3.5-min segment whereby footage of her unexpressive conducting was inserted during the ensemble performance portions of the rehearsal. The verbal instruction portions were unaltered, and for the entire alternative version, the audio track from the original, expressive conducting segment was used. We mixed and balanced the audio levels from the three audio sources (i.e., the two main cameras and the separate audio microphone), and we adjusted the sound levels as needed over the course of the segments so that both the conductor and the ensemble could be clearly heard. Transitions between varying camera angles (and for the altered segment, transitions between the verbal instruction and the unexpressive conducting episodes) were made to be as seamless as possible. Camera angles were consistent between altered and unaltered versions; for example, if the view switched from the left camera to the right camera at a particular point in the rehearsal for the altered version, then the same camera angles were used at that same moment in the unaltered version.
To verify that the altered rehearsal video would otherwise appear to be natural to viewers (i.e., not constructed from multiple recordings, as was actually done), we showed the segment to a panel of five graduate music students. We did not explain the purpose of the study to them or the manner in which we created the video segment. We asked them to comment freely about the production value of the video segment and about the rehearsal itself. None of them mentioned anything about the segment seeming artificially constructed.
In total, three 3.5-min segments were created: (a) experimental conductor conducting expressively (hereafter abbreviated as “expressive”), (b) experimental conductor conducting unexpressively (i.e., the altered version, “unexpressive”), and (c) “control” conductor. All were rendered as QuickTime video files of a size and quality suitable for online viewing.
Procedure
An Internet link was provided via email to all prospective evaluators. Those choosing to participate were instructed to click the link, whereupon they were led to Qualtrics, an online survey system. Institutional Review Board–approved consent information appeared on the first screen, followed by a screen on which the evaluation task was described: You will view two, 3.5-minute segments of two different conductors leading a concert band rehearsal. After each segment, you will be asked to evaluate the quality of the conductor’s teaching by providing a rating on eight different statements related to rehearsal effectiveness [see Table 1], and by providing open-ended comments about the rehearsal. Please view each segment in its entirety before clicking through to the evaluation screen. When you give your responses, treat each criterion as a separate and distinct evaluation item, and offer your response ONLY in consideration of that item.
Means (and Standard Deviations) of Evaluation Items of Experimental Conductor in Expressive (Unaltered) and Unexpressive (Altered) Conditions.
Note. Evaluations are on a 10-point scale, from 1 (poor) to 10 (excellent). Order 1 = expressive, then control; Order 2 = control, then expressive; Order 3 = unexpressive, then control; Order 4 = control, then unexpressive.
Italicized text represents modifications from or additions to the Bergee (1992) Rehearsal Effectiveness Scale for purposes of this study. (Text was not italicized as it appeared to participants.)
Participants then viewed one of four possible combinations of two of the video segments: (a) expressive, control; (b) control, expressive; (c) unexpressive, control; or (d) control, unexpressive. (These labels were not shown to participants.) The particular combination a given participant would see was determined randomly by the survey system to ensure an even distribution of viewing orders among all participants.
To measure our participants’ perceptions of teacher effectiveness, we adopted eight items from Bergee’s (1992) Rehearsal Effectiveness Scale for use as dependent measures in our study. We deliberately chose items that directed evaluators’ attention to verbal and instructional aspects of teaching rather than to aspects related to gesture and conducting. Following their viewing of each of the two rehearsal segments, participants were asked to rate the conductor on these items using 10-point Likert-type scales anchored by poor (1) and excellent (10). An open-ended response item followed the rating scales, with the instruction, “in the space below, please offer any other thoughts you may have about the effectiveness of the conductor’s teaching as seen in this rehearsal.” At the conclusion of the evaluation task, participants were asked to give information about their gender, age, year in school, degree program, major instrument, and years of performing in a conducted ensemble. Average participant completion time for the survey procedures was 12 min, 38 s. One hundred seventy-four people began the online survey, but we eliminated responses from 38 people due to incomplete responses or insufficient time spent with the survey procedures as revealed in the Qualtrics data report.
Results
We inspected mean evaluation scores on each statement for the experimental conductor in both expressive and unexpressive conditions and in orders where the experimental conductor was viewed either prior to or following the control conductor. Results are reported in Table 1. For both orders, evaluations for all of the eight statements were higher in the expressive condition than the unexpressive condition. The largest such difference between conditions was for the statement, “Demonstrates ability to motivate students,” with a difference of 1.66 points when viewed prior to the control conductor and 1.03 points when viewed afterward.
We then subjected the evaluation scores for the experimental conductor, on all criteria and in both conditions, to a principal components analysis procedure with varimax rotation. All evaluation criteria loaded positively onto a single factor (i.e., there was only one factor with an eigenvalue greater than 1), with loadings ranging from .781 (for the criterion “Demonstrates ability to motivate students”) to .905 (for “Appears organized and efficient”), together explaining 69.2% of the total variance in the model. This single-factor solution was also confirmed by visual inspection of the scree plot. Given this large amount of shared variance among the evaluation criteria, it therefore seemed prudent to proceed with further analysis by averaging scores on the eight separate criteria into a single rehearsal effectiveness variable.
We performed a Box-Cox power transformation (Box & Cox, 1964; Osborne, 2010) on the averaged scores to achieve satisfactory normality and homoscedasticity in the data set. We then conducted a two-way analysis of variance (ANOVA), with condition and order serving as between-subjects factors. Results indicated a significant effect for condition, F(1, 130) = 19.861, p < .001, η p 2 = .133; but not order, F(1, 130) = 1.377, p = .243, η p 2 = .010; or the interaction of condition and order, F(1, 130) = 0.274, p = .602, η p 2 = .002. These data are depicted in Figure 1.

Mean evaluation scores (average of eight criteria, each scored from 1 = poor to 10 = excellent) according to expressivity condition and order. Error bars represent standard error.
Evaluation scores for the control conductor were not of interest in our current study, nor were comparisons between her scores and those for the experimental conductor. However, to further confirm the validity of the differences we saw between conditions for the experimental conductor—that is, to help establish that the evaluation scores were reflective of differences between the conditions and not of inherent differences between the participants who evaluated them—we ran similar ANOVA tests for the control conductor’s averaged evaluation scores (even though her video was unaltered between conditions, with her always being seen conducting expressively), again with a Box-Cox transformation. We found no significant differences for condition, F(1, 130) = 0.908, p = .342, η p 2 = .007; order, F(1, 130) = 0.326, p = .569, η p 2 = .003; or for the condition-order interaction, F(1, 130) = 0.117, p = .733, η p 2 = .001.
To further explore relationships between expressivity conditions and each of our criteria, we performed rank-biserial correlations on the experimental conductor’s evaluation scores. Results are reported in Table S1 in the online version of the article. Seven of the eight criteria (all except “Maintains a professional demeanor”) showed significant, small, positive associations with expressivity.
We followed analysis protocols from related studies (Montemayor, 2016; Montemayor & Moss, 2009; Silvey & Montemayor, 2014) in our examination of the participants’ written comments about the effectiveness of the experimental conductor’s teaching. We began our preliminary analysis by coding comments according to two broad categories, reflecting our independent and dependent variables, namely, gesture and verbal instruction (i.e., nongestural aspects of teaching). Upon inspection of the data set and after discussion, we added two further emergent categories: leadership (i.e., general comments about perceived personal qualities and instructional capacities) and musicianship (i.e., inferential comments regarding the overall musical skills of the conductor). We further identified each comment within these categories as being either positive or negative. Together, these four categories provided for a parsimonious interpretation of the complete set of participant comments pursuant to our research question. For each comment, multiple categorizations were possible. The primary author and a graduate assistant independently examined 25% of the participants’ responses (with participant group assignments masked to us throughout the process); interobserver agreement on these categorizations was 92.4%. We then independently examined the remainder of the data set and resolved any remaining disagreements through discussion. We counted 208 identifiable comments according to these categories. Results are presented in Table S2 in the online version of the article. Overall numbers of comments were similar between conditions, as were proportions of comments addressing general leadership qualities of the conductor. As expected, participants’ comments regarding gesture were more positive in the expressive condition and more negative in the unexpressive condition. However, negative comments regarding verbal behavior were nearly three times as frequent in the unexpressive condition (17 such comments vs. 6 in the expressive condition). Also, viewers of the expressive condition provided five times as many positive comments about the conductor’s general musicianship (15) than those who viewed the conductor conducting unexpressively (3).
Discussion
We found that when our experimental conductor was viewed conducting expressively, participants rated several measures of teaching effectiveness higher than when that same individual demonstrated unexpressive conducting even though the teaching remained constant. We speculate that participants were unable to separate nonverbal conducting behaviors (e.g., expressive gesture, eye contact, and facial expression) from verbal instructional processes in their ratings of the experimental conductor. Even in our printed directions to participants, we explicitly indicated that they should evaluate each statement independent of any other item. This finding indicates that these observers concurred with the pedagogical belief that “good conducting is good teaching” (Harris, 2001; Seaman, 2013; Sidoti, 1990). It appears that the presence of expressive conducting behaviors interspersed among episodes of verbal instruction influenced perceptions regarding the efficacy of those verbalizations. This finding may give credence to the idea that undergraduate conducting courses should include the discussion, teaching, and student demonstration of both nonverbal and verbal conducting skills—a curricular design that is atypical of most undergraduate conducting sequences in which there are often separate course offerings for learning conducting and rehearsal behaviors (Manfredo, 2008).
One outcome of particular interest was that our participants perceived rehearsal pacing to be more effective in the expressive conducting condition than the unexpressive condition when in actuality, the pacing remained unchanged. This finding may be congruent with other research indicating that pacing is likely related to perceived incongruities in focus of attention (Madsen, 1997). Perhaps when there is a discrepancy between what a conductor says versus what he or she shows, ensemble members are more likely to lose concentration during the music-making process. The admonition to “show more and talk less” that is often given by conductor educators to novices may serve as more than anecdotal advice as perceptions of pacing (and perhaps other teaching behaviors) are indeed enhanced through the demonstration of conducting skills alone (see also Duke, Prickett, & Jellison, 1998). Future researchers should consider experiments in which specific teaching behaviors such as feedback, pacing, and sequencing are manipulated independently in conjunction with expressive and unexpressive conducting behaviors.
We believe that our quantitative findings are similar with other studies indicating the dominance of visual over auditory information when making decisions about musical performances (Davidson, 1993; Juchniewicz, 2008; Nápoles & Silvey, 2017; Silvey & Fisher, 2015; Tsay, 2013, 2014). Although we did not ask our participants in this study to rate ensemble performance, it is interesting that all of the teaching variables were rated higher by participants when the rehearsal was conducted expressively rather than unexpressively. Given that ensemble members have been found to enjoy the music and their conductor more when the conducting was expressive versus unexpressive (Price & Winter, 1991; Silvey & Koerner, 2016), our results seem to suggest that conducting expressivity influences beliefs regarding overall teaching effectiveness, a finding that may indicate that conducting and teaching are viewed as inseparable in the rehearsal process.
Participants’ written responses seem to affirm and extend the results from their numerical evaluations. Negative comments about the conductor’s verbal behavior were much more frequent in the unexpressive condition even though what the participants observed in that regard was identical to what those who were shown expressive conducting had seen. Perhaps in the absence of expressive gesture, observers perceived instructional deficits in terms of what the conductor should or should not have verbalized from the podium. General comments about the conductor’s leadership skills were rather similar between conditions, perhaps reflecting the lack of a significant relationship on the “maintains a professional demeanor” criterion; this might also be indicative of participants’ attempts to articulate thoughts that accurately discriminated among various facets of rehearsal instruction. Perhaps most interesting is the comparatively high frequency of positive comments about the conductor’s musicianship in the expressive condition—comments that were otherwise almost entirely absent in the unexpressive condition. Expressive gesture, like skilled performance on an instrument or voice, appeared to these observers to be a manifestation of general musical capacities. Evaluators’ tendencies to extrapolate from observed phenomena and infer global teacher characteristics underscore the importance for prospective ensemble conductors to appear expressive while rehearsing.
These results also affirm the primacy of delivery skills as a central component of perceived instructional effectiveness in music settings. Researchers in previous studies have found that teaching episodes featuring high-affect delivery are evaluated as being more effective than ones with low affect even in the presence of poor or inaccurate content (Hamann et al., 2000; K. Madsen, 2003). Moreover, perceptions of student progress are influenced by teacher delivery, with even limited moment-to-moment student achievement being perceived as reflecting greater or lesser progress according to the behavior of the teacher (MacLeod & Nápoles, 2014; Napoles & MacLeod, 2013). In the current study, we manipulated expressive components of conducting segments only and not the quality of the verbal/instructional segments or the ensemble’s performance. Our finding that evaluations of nongestural aspects of teaching are influenced by expressive conducting is congruent with those from earlier studies and further demonstrates the interrelatedness of discrete elements of teacher behavior within music teaching and learning settings.
It should be noted that we examined perceived effectiveness only, as observed over a relatively brief span of time, and that the evaluators were people other than those in the ensemble being conducted. We are unaware of other studies that have investigated longer-term effects of observed musical gesture as seen in instructional contexts rather than performance settings. Again, however, positive effects on instrumental performance have been found with young musicians practicing conducting movements over the course of several days or several weeks (Cofer, 1998; Kelly, 1997; Thompson, 2012). Given these results (and those from the current study), it seems possible that observed expressive gesture could improve musical performance not only through commanding heightened attention to verbal instruction but also by means of a teacher’s visual modeling of a desired aural response (Acklin, 2009; Mayne, 1992; Selvey, 2014; Sousa, 1988). Also, in recruiting participants, we drew from a general population of college musicians, ranging from non–music majors (the majority of participants) and voice majors playing in auditioned bands to graduate music students. Given that numerous music education researchers have identified various training- and experience-based differences in the evaluation of instruction (Duke, 1987; C. K. Madsen & Duke, 1985a, 1985b, 1987; K. Madsen & Cassidy, 2005; Standley & Madsen, 1991; Yarbrough & Hendel, 1993; Yarbrough, Price, & Hendel, 1994) and in evaluating the results of instruction (Montemayor, 2016), future investigators could determine if the perception of teaching effectiveness (or even the perception of discrete teacher behaviors) in the presence or absence of conductor expressivity varies according to these factors. Results from those earlier studies may suggest that more experienced musicians could have greater expectations for expressive conducting, but whether or not they are better able to offer evaluations that distinguish between conductor gesture and teacher verbiage remains a topic to be explored.
Our findings also invite new questions regarding the perception of conducting in teaching and learning contexts. When do musical gesture and musical instruction begin to seem inseparable to observers, and how are perceptions further influenced by the interactions among these components and the musical contexts in which they occur? Do observers respond more to conducting gesture or to facial expressions (Wöllner, 2008), and to what extent are observers’ expectations a matter of congruence among conductor gesture, conductor facial expressions, musical affect, and/or teacher verbiage? The positive effects of “nonverbal immediacy” teacher behaviors (including expressive gesture) appear to be recognized across cultures (McCroskey et al., 1995), as they did for participants in this study. Answers to these and related questions would contribute to a greater understanding of how students perceive and ultimately respond to gesture as a part of their instructional experiences in music settings.
Supplemental Material
DS_10.1177_0022429419835198 – Supplemental material for Conductor Expressivity Affects Evaluation of Rehearsal Instruction
Supplemental material, DS_10.1177_0022429419835198 for Conductor Expressivity Affects Evaluation of Rehearsal Instruction by Mark Montemayor and Brian A. Silvey in Journal of Research in Music Education
Footnotes
Acknowledgements
We thank Aaron Pendergast of Static Age Productions (Denver, Colorado) for his generous assistance in the video recording and editing process. We also thank Yolanda Chatwood, Jennifer L. Grice, and Sarah Romero, graduate students at the University of Northern Colorado, for their assistance with this study.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Author Biographies
References
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.
