Abstract
This study investigates whether experiential (informal/non-formal) sight-reading practice has an impact on sight-reading skills of classical university-level pianists. Data collected from 80 self-report practice diaries of students participating in trials of new sight-reading pedagogies were analysed to evaluate the effect of experiential sight-reading practice on improvement in sight-reading performance, focusing on the overall duration of sight-reading activities accumulated over the 10-week period and different patterns across this time. Data were subjected to bivariate regressions and bivariate correlations analyses. Results showed that improvement in sight-reading performance on three out of the four key indicators (beat adjustment, extra notes and missing notes) could not be predicted by participants’ overall hours of experiential practice. Yet the total number of hours engaged in sight-reading practice did predict an improvement in participant RMS accuracy (average timing errors for each correct note played). Findings also revealed no consistent pattern of practice across the 10 weeks that was associated with any of the four performance indicators. Implications for education include the need to embrace multi-faceted approaches to sight-reading training by focusing on skills contributing to expert sight-reading and their development, individually and in combinations. Rhythm and pitch training, development of pattern recognition and prediction skills, and collaborative playing activities have the potential to enhance sight-reading skills of music undergraduates and have positive flow-on effects for their preparation for 21st century music careers.
Keywords
The ability to learn new repertoire quickly is an essential skill for any performing musician. While contemporary, popular and jazz performers tend to rely on improvisatory and aural skills in their repertoire learning, for classical musicians the ability to read a score and rapidly translate notation into playing is of paramount importance. The speed and effectiveness of learning new repertoire depends largely on the ability to process the musical notation accurately during the initial reading, i.e., the ability to sight-read.
Despite years of research into sight-reading, the debate about whether fluent sight-reading is a natural talent or a skill that can be developed has not been resolved (Zhukov, 2006). Many studies have focused on behavioural and cognitive processes during sight-reading to identify predictors of expert performance in this area. For example, W. B. Thompson (1987) showed that sight-reading ability in flautists was significantly correlated with eye–hand span, and Kopiez and Lee (2006) demonstrated that for pianists trilling speed was a predictor. Meinz and Hambrick (2010) suggested that working memory plays an important role in sight-reading expertise, and Brodsky, Henik, Rubinstein, and Zorman (2003) discovered that nearly half of their participants were able to generate auditory imagery of silently read musical notation. These studies imply that musicians’ inherent physical and/or mental capacities have the greatest impact on fluent sight-reading. However, this view has been challenged recently by Mishra (2014a), who demonstrated in her meta-analysis of 92 sight-reading studies that stable characteristics such as musical aptitude, IQ and personality correlate less strongly with sight-reading expertise than certain musical skills that could be improved with targeted practice.
A number of authors have proposed various strategies to enhance sight-reading skills, however these recommendations are often based on personal opinions and experiences, rather than rigorous research. For example, Wristen (2005) suggested that solfège, breaking the score into chunks when reading and collaborative playing activities might be beneficial for developing sight-reading ability. S. Thompson and Lehmann (2004) highlighted the importance of knowledge of musical periods and characteristics of different styles, and Hodges and Nolker (2011) recommended training in pattern recognition and rhythm to improve prediction skills and rhythmic accuracy in sight-reading. To improve visual processing of the score during sight-reading, Lehmann and Kopiez (2015) suggested that piano students should play without looking down at their hands, and that attempting to “fake” playing in the relevant style, based on the knowledge of its features, could reduce interruptions to the flow of pianists’ sight-reading performance. However, few studies investigating sight-reading have been conducted in controlled environments, using large samples, employing lengthy treatments, and applying statistical analyses of results, or have been replicated by other researchers. Thus the available research has provided only limited implications for sight-reading pedagogy. One exception is a study conducted by Zhukov, Viney, Riddle, Teniswood-Harvey, and Fujimura (2016), who demonstrated that sight-reading skills in advanced pianists could be improved using a combined pedagogical approach based on findings of a large-scale controlled study.
Research has demonstrated that both the quantity and the quality of music practice affect the development of performance expertise (for a review, see Zhukov, 2009). To achieve an expert level of musical performance a minimum of 10 years of deliberate practising is required (Ericsson, 1996), with practice sessions increasing in frequency and duration during primary, secondary and higher education (Jorgensen, 2004; Sloboda, Davidson, Howe, & Moore, 1996), and practice strategies being enhanced through the use of meta-cognitive approaches (Hallam, 1997; McPherson, 2005), mental practice and score analysis (Fine, Goldemberg, Wise, & Bravo, 2015), and application of systematic strategies, organisation, recordings, and metronome use (Hallam et al., 2012). However, recently, Hambrick, Altmann, Oswald, Meinz, and Gobet (2014) have challenged the accepted views on the importance of deliberate practice, suggesting that other factors such as starting age, memory capacity and genes play a greater role in achieving expert performance in all major domains. In terms of memory capacity, long-term memory models suggest that privileged access to long-term memory is the result of years of training (Ericsson & Kintsch, 1995) and music practice has a positive effect on the development of working memory during childhood and adolescence (Nutley, Darki, & Klingberg, 2014).
Despite the belief commonly held by many teachers, students and music professionals that simply partaking in informal sight-reading activities will improve sight-reading performance, little evidence exists to support this premise. A recent review of over 60 years of studies into sight-reading (Mishra, 2014a) highlighted that effective interventions to enhance sight-reading include components such as aural skills training, composition, improvisation, and solfége. For example, Waters, Townsend, and Underwood (1998) showed through a series of laboratory experiments that “auditory skills and prediction skills are important factors underlying skilled sight-reading over and above basic pattern-recognition skills” (p. 147). The study found that expert sight-readers processed larger chunks of music and made better use of contextual information such as key signature and harmony. This finding is supported by the results of Fine, Berry, and Rosner (2006) who found that pattern recognition and harmonic prediction were an essential element of sight-singing ability. Kopiez and Lee (2008) highlighted the role of inner hearing in the development of piano sight-reading excellence since it functions “as an additional information channel that would be useful for anticipatory information processing” (p. 56). While pattern recognition and prediction capabilities are factors contributing to expert sight-reading, these skills are often developed during composition and improvisation studies, but not necessarily during sub-optimal, experiential sight-reading practice.
Over the past 20 years technological advances in eye-tracking and brain-monitoring equipment have helped us to understand the physiological and cognitive processes involved in sight-reading and pointed to new directions for developing effective pedagogical interventions. Utilising eye-tracking equipment, the behavioural research into sight-reading has largely focused on piano sight-reading (e.g., Drai-Zerbib, Baccino, & Bigand, 2012). Typical findings show that, in comparison to novices, expert pianists take fewer and shorter glances when sight-reading due to their ability to process large groups of notes quickly (Waters, Underwood, & Findlay, 1997). Wurtz, Mueri, and Wiesendanger (2009) is one of the few studies that investigated non-piano sight-reading. These researchers measured the eye–hand span (i.e., the time lapse between reading and executing the notes) of violinists during sight-reading of two stylistically similar examples by Corelli and Telemann. The reported results support earlier findings in pianists, with superior prediction skills being linked to violinists’ experience but also highlighting the importance of tempo, rhythmic and melodic characteristics of the work being played, and how these musical features are interpreted by the players as factors impacting on fluent sight-reading. The eye-tracking studies of sight-reading confirm the importance of developing pattern recognition and prediction capacities in students to enhance their sight-reading skills.
Recent advances in brain-monitoring technology have provided new insights into how pitch and rhythm are processed during sight-reading. For example, D’Anselmo, Giuliani, Marzoli, Tommasi, and Brancucci (2015) validated the commonly held view that pianists experiencing difficulties during sight-reading frequently omit playing some of the notes in the left hand. The study found hemispheric brain asymmetries during sight-reading of musical notation, with a bias towards the right hand when playing both hands together. The long-held belief that auditory skills play an important role in expert sight-reading has been validated and also challenged by brain research. Gunter, Schmidt, and Besson (2003) used electroencephalogram (EEG) to measure classical musicians’ responses to diatonic violations of notated familiar melodies presented as scores only and scores with recordings. The study found that reading and listening to the same errors in music does not trigger the same electrophysiological activations in the brain. This suggests that reading of music and listening to music may have less in common neurologically than previously thought. Despite this difference, brain research has also shown that when sight-reading, musicians tend to process the visual score into an auditory prompt, store this in short-term memory and compare it to the sound they are producing immediately following (Simoens & Tervaniemi, 2013). The EEG evidence supports the theories that music sight-reading combines cross-modal processing and short-term memory functions previously flagged by behavioural research (Kopiez, Weih, Ligges, & Lee, 2006).
Fluent sight-reading involves the processing of musical pitches and their durations simultaneously. These musical concepts are taught concurrently when young musicians are starting instrumental learning. However, brain research has demonstrated that pitch and duration are not processed jointly (Jerde, Childs, Handy, Nagode, & Pardo, 2011). This finding suggests that it might be beneficial to focus singly on rhythm training without pitch and pitch processing without rhythm to mediate weaknesses in sight-reading. Mishra (2016) supports this notion through her analysis of 46 rhythmic studies of sight-reading conducted between 1954 and 2012. She found strong evidence that “interventions that included a counting system, movement, or rhythmic drill effectively enhanced rhythmic sight-reading” (p. 9). However, the notion that sight-reading accuracy depends on the ability to maintain internal pulse was not supported by Farley (2014), who nonetheless found a significant correlation between rhythm reading and time-keeping. From her review of 21 melodic studies of sight-reading Mishra (2016) concluded that melodic sight-reading might be enhanced with collaborative playing and general instrumental training. Zhukov et al. (2016) documented a similar approach that combined collaborative playing activities with rhythm training and developing understanding of different styles, utilising a combined training program that impacted positively on sight-reading of university-level pianists.
Many of the studies reporting improvement in sight-reading have been conducted in “uncontrolled environments”, with “contradictory findings accepted without comment or analysis” (Mishra, 2014b, p. 149). A recent study by Rosemann, Altenmüller, and Fahle (2016) focusing on the effect of practice on pianists’ eye-span is yet another example of limitations of experimental research in the area: this particular trial involved only nine participants, 30 minutes of practice, uncomfortable eye-tracking equipment and loss of data, resulting in a predictable conclusion that short rehearsals do not improve eye–hand span, and that this characteristic is developed over many years of practice and is unique for each pianist.
The debate on how best to train sight-reading skills is still ongoing. The lack of research evidence on validated pedagogical approaches for improving of sight-reading means that studio teachers tend to rely on their own experience, intuition and anecdotal evidence (Gudmundsdottir, 2010; Hodges & Nolker, 2011). This often results in trusting the old adage that practice makes everything perfect, including sight-reading. When music teachers refer to practising sight-reading, they often mean simply providing students with sight-reading experiences and employing non-goal-directed strategies that do not necessarily improve the intended skill. However, recent research has highlighted the need to distinguish between mere experience as a non-directed activity and deliberate practice (Platz, Kopiez, Lehmann, & Wolf, 2014). Controversially, Meinz and Hambrick (2010) suggested that while deliberate practice is necessary, it is not sufficient to explain differences in sight-reading skills. They offered no alternative explanations for the sight-reading variances in their results, and some of the statistical procedures used in their study were criticised by Platz et al. (2014). The question of what could possibly contribute to improvement in sight-reading and explain individual differences in this skill remains unanswered.
The aim of this article is to investigate whether experiential (informal/non-formal) sight-reading practice does lead to improvement in sight-reading skills. This will be achieved through the analysis of self-report data of a large sample of advanced pianists.
Method
Participants
The main study measured the impact of three single-focus training programs (accompanying, style, and rhythm) on sight-reading of university-level pianists versus control (for details, see Zhukov, 2014). To match the level of pianistic skills and experience, the selected participants (N = 100) were of Grade Eight and above level, which is classified by the Australian Music Examination Board as advanced. Each of the treatment groups (n = 25 × 3) undertook 10 weeks of training that consisted of materials focused on one specific area: the accompanying group practised two short pieces per week accompanying a partner; the rhythm group performed exercises that involved complex rhythms and guided analysis; and the style group played two short Baroque/Classical pieces and analysed their structure and harmony. The control group (n = 25) was given no special tasks.
Eighty out of the 100 pianists submitted diaries of their sight-reading practice over the 10 weeks of training, relatively evenly distributed amongst the treatment and control group participants: 19 control, 21 accompanying, 21 rhythm and 19 style diaries (see Table 1 for demographics).
Sample demographics.
Procedure
The participants’ sight-reading was tested on three unfamiliar stimulus pieces, each one-page long and of intermediate difficulty, before and after a 10-week training program. The same test pieces were used for pre- and post-tests, a procedure employed in previous research (Betts & Cassidy, 2000; Smith, 2009). None of the participants recognised the repeated materials during post-tests, confirming that participation in the training programs and the amount of music generally encountered during the 10 weeks between the tests were sufficient to prevent recall. One-minute perusal time was given for each test piece, after which a medium tempo was established by the metronome click that continued throughout playing which was recorded using Cubase software.
The MIDI files of playing were analysed by software specifically designed for the study. The software was developed to accommodate the stop-and-start nature of playing typical during student sight-reading. An algorithm was developed to tolerate time-domain errors that were not recognised by other systems of analysis, matching note events between participant performance and an ideal. A specific time-window for scoring of correct notes was set for each test piece, with accuracy margin set to the length of the shortest beat subdivision as measured in MIDI ticks, i.e., in Test Piece 1 the shortest beat length was 240 and, therefore, the accuracy margin was also set at 240. The time-domain errors were counted as beat adjustments and correct pitches played were matched to an ideal performance through application of time-domain shifts, generating four numerical scores: two on pitch accuracy and two on rhythm accuracy. These four performance indicators were: (1) beat adjustment (number of interruptions to flow, skipped or missed beats); (2) extra notes (additional pitches played that were not notated on the score); (3) missing notes (pitches that were notated on the score but were missing in performance); and (4) RMS accuracy (average timing errors for each correct note played).
Diary data
In addition to pre- and post-tests of sight-reading, the participants were required to keep a practice diary for 10 weeks and enter the duration of their sight-reading activities on a daily basis. While self-report data may result in socially desirable responding where people over-report the number of minutes engaged in sight-reading, participants were encouraged to document accurately the amount of time spent in sight-reading activities and not to exaggerate. To this end an example provided on the diary’s inside page showed no entries on some days of the week. Further, since all interactions between the researcher and the participants were strictly study-related and not in the context of a teacher–student power relationship, it was assumed that the information gathered was as accurate as possible under the given circumstances.
This paper reports on the analyses of practice diary data in relation to improvement in participant sight-reading performance. The accumulated total number of minutes spent sight-reading over the 10-week period was calculated for each participant. The effect of experiential practice on improvement in sight-reading performance (i.e., change in scores from pre-test to post-test on the four performance indicator categories) was evaluated, focusing on the overall amount of sight-reading practice over the 10 weeks and different practice patterns across this time. Statistical analyses were carried out using the statistical program SPSS Version 21.0.
Results
The study involved a 2 (pre–post: pre-training, post-training) × 4 (training condition: control, accompanying, rhythm, style) × 3 (test piece: test piece 1, test piece 2, test piece 3) mixed factorial design. In this, pre–post and test piece were within participants factors, while training condition was the between participants factor. However, the focal predictor was the number of hours of practice participants engaged in every week over the 10-week training period, which was summed to give a total value. The key outcomes were the four musical proficiency measures in the form of (1) beat adjustment, (2) extra notes, (3) missing notes, and (4) RMS accuracy, where lower values on each were indicative of greater proficiency (i.e., fewer errors).
First, difference scores were created by subtracting participants’ post-training from pre-training score on each of the musical proficiency indicators of beat adjustment, extra notes, missing notes, and RMS accuracy, averaged across the three example pieces. Since each test piece was not in itself of focal interest in this study and the preliminary analyses showed no clear pattern with regard to this variable, a broader view of examining errors across the three example pieces was adopted as this provides a more stable index of mean participant improvement (i.e., average reduction in specific type of error across the three pieces) on these four key performance outcomes from their baseline measure.
Initial checks of the data revealed the presence of two outliers on the total number of practice hours. Both possessed standardised scores (i.e., z = 3.60 and z = 7.23) exceeding the ±3.29 threshold recommended by Tabachnick and Fidell (2013), equating to these scores significantly differing from the mean of the rest of the sample at p < .001. Given that analyses with and without these two participants produced substantively different results and both represented exceptionally motivated individuals who had undertaken excessive amounts of self-initiated practice from the control group, these two participants were deemed not to be representative of the sample and were removed from the dataset.
Comparison of training conditions
The next step was to check for any potential differences in practice hours that may have occurred between the various training conditions. Once the data from the two outliers was eliminated, a one-way analysis of variance revealed that the total practice hours participants engaged in across the 10 weeks did not differ significantly amongst the training conditions, F(3, 74) = 2.51, p = .066, η2 = .09. Given that no systematic training group difference was found between the control (M = 557.06, SD = 424.47), accompanying (M = 377.81, SD = 261.63), rhythm (M = 320.48, SD = 194.26), and style conditions (M = 401.68, SD = 164.88) on the total number of practice hours undertaken, indicating similar number of hours of practice across the 10 weeks for participants within each of these groups, it was deemed appropriate that analyses collapse over training conditions rather than examine each separately.
Impact on performance outcomes
A series of four bivariate regressions were conducted to establish whether the total number of practice hours engaged in over the 10-week period by participants could predict improvement in the four key performance outcome measures. Findings revealed a higher number of total hours spent practising sight-reading significantly predicted performance improvement in the form of RMS accuracy (R2 = .05, β = .23, p = .046). However, the remainder of the bivariate regressions showed no significant improvement in participant beat adjustment, extra notes, and missing notes due to the total number of hours spent practising across the 10 weeks, all p values > .087.
Patterns of practice
To explore whether unique patterns existed within the 10-week time span indicative of critical practice periods for the key musical performance outcomes, bivariate correlations were also performed to assess whether there was a relationship between the hours of practice undertaken by participants each week and the four performance indicators of beat adjustment, extra notes, missing notes, and RMS accuracy. Since only one out of 40 correlations (2.5%) was significant, this could easily signify a Type I error. Even taking into account the four marginally significant relationships found, there was no clear nor systematic pattern that any practice week was more influential than the others, as each appeared to be equally ineffectual on its own.
Given the combined results of the bivariate regressions and bivariate correlations (broken down by practice week), it would appear that improvement in sight-reading performance was not associated with participants’ patterns of practice over the duration of the 10 weeks. However, the bivariate regression analyses found that one of the four performance indicators – RMS accuracy – was predictable based upon the cumulative overall hours of practice, while the performance of beat adjustment, extra notes, and missing notes were not.
Discussion
The aim of this study was to evaluate impact of experiential sight-reading practice on improvement in pitch and rhythm accuracy of sight-reading performance in advanced pianists. In particular, structuring a controlled environment, using a large sample, and implementing data collection over a relatively long period of time were strategies employed here to address inconsistencies in previous research as highlighted by Mishra (2014a, 2014b, 2016). The pre- and post-test performance data were analysed using software, thus eliminating human bias and error in the evaluation of playing. Due to large amount of data it was possible to execute statistical analyses (bivariate regressions and bivariate correlations) to explore any possible connections between the sight-reading activities and improvement in any of the four performance indicators.
The results show that total number of minutes spent in experiential practice across the 10-week period did not appear to improve musical performance on three of the four key proficiency indicators that were examined: extra notes, missing notes, and beat adjustment. However, the cumulative practice did predict improvement in RMS accuracy (i.e., reduction of average timing errors for each correct note played). The limited findings do not lend support to the long-held belief that partaking in informal sight-reading activities will improve overall sight-reading performance, and instead take sides with previous research which suggested that such a view is based on largely anecdotal evidence (Gudmundsdottir, 2010; Hodges & Nolker, 2011).
No consistent practice patterns emerged with regard to amounts of practice or the training condition in terms of impacting on sight-reading performance. The lack of advantageous practice approaches identifiable over the 10-week period suggests that beneficial strategies for improving sight-reading lie beyond experiential practice as highlighted by meta-analyses of previous research (Mishra, 2014a).
Why did informal sight-reading practice have such a small impact on sight-reading performance? The substantial volume of research on instrumental and vocal learning undertaken over the past 20 years has demonstrated to teachers and students that simply accumulating many hours of practice in itself will not guarantee high achievement, and that musical expertise is achieved through a complex interaction of many factors (McPherson & Williamon, 2016). Attaining expert level of playing also requires implementation of a variety of deliberate practising strategies that have been shown to be effective by research, including: combining the chunking of difficult passages into manageable practice sub-sections with a run through the entire piece (Jorgensen, 2004); mental rehearsal (Connolly & Williamon, 2004; Fine et al., 2015); and reflection and self-evaluation of practising, leading to meta-cognition (Hallam, 1997; McPherson, 2005). However, in the area of sight-reading we are as yet to observe any changes in practice strategies, with many teachers and students still clinging to the obsolete idea that simply engaging in sight-reading activities will improve fluency and accuracy of this skill.
What could improve sight-reading if informal sight-reading activities do not? Sight-reading is a very complex skill that involves high order cognitive and motor processing, therefore utilizing targeted methodologies that focus on a number of specific factors contributing to fluent sight-reading as identified by research (Mishra, 2016) and developing these skills individually and in combinations, will be much more likely to lead to an improved overall performance on sight-reading tasks.
Sight-reading involves simultaneous pitch and rhythm processing, therefore it follows that developing these skills through deliberate practice approaches could be beneficial. Recently, brain research (e.g., Jerde et al. 2011) showed that rhythm and pitch have unique neural signatures during auditory processes, suggesting that focusing on rhythm and pitch processing singly could be more beneficial than addressing these issues simultaneously as is typical during sight-reading exercises. Such an approach is supported by sight-reading research that investigated these concepts individually (Mishra, 2016).
The separation between aural training and instrumental/vocal learning continues at every level of music education: beginner students tend to attend theory/ musicianship classes that are typically separate from their weekly practical lessons; in primary and secondary education generic music skills are developed during classroom lessons rather than incorporated into performance situations such as band and choir rehearsals; in higher education students participate in aural musicianship classes, but tend not to transfer the skills learnt there to instrumental performance. Therefore, it is important to incorporate aural and rhythm training into individual lessons at every level of instrumental/vocal learning.
A pilot study by Green (2012) showed that playing-by-ear activities during instrumental lessons facilitate the development of pitch-sense, rhythm-sense, and enhance learning strategies among adolescent musicians. A follow-up study by Varvarigou (2014) demonstrated that in addition to developing students’ aural skills these strategies resulted in greater confidence of teenage musicians in their instrumental playing. Such approaches are more beneficial to improving pitch and rhythm processing than mere error corrections and ensure that underlying concepts are understood rather than simply copied by students.
Research has highlighted the role of pattern recognition and prediction skills, and inner hearing in fluent sight-reading (Kopiez & Lee, 2008; Waters et al., 1998) and sight-singing (Fine et al., 2006). Non-goal-oriented sight-reading activities cannot develop these skills. The strategies to enhance pattern recognition and prediction skills include harmonic and structural analyses, often implemented in practical manner in improvisation and through a more theoretical methodology in composition. To apply these ideas in individual lessons teachers could, for example, include short improvisatory exercises that will help students focus on particular melodic and harmonic patterns present in the work being studied, and, in addition to analysing the overall structure and harmony of the work, ask students to compose a miniature piece based on the form of the work they are learning.
While research has demonstrated that musicians actively involved in collaborative playing activities, such as accompanists and répétiteurs, have superior sight-reading skills (Lehmann & Ericsson, 1996) and recommended the inclusion of such strategies to benefit the development of sight-reading (Wristen, 2005), the evidence that collaborative playing does enhance sight-reading skills is beginning to emerge (Zhukov et al., 2016). Teachers typically play duets with beginner students to produce a bigger sound and create a more enjoyable atmosphere during lessons. This approach tends to taper off during adolescence as students gain more performance skills, with lessons becoming more focused on complex repertoire and the development of technique. It is important to maintain collaborative playing as a component of instrumental lessons through secondary and higher education by encouraging students to play duets, accompany their peers, and play chamber music. Collaborative performance requires musicians to keep going and, therefore, improves overall flow by limiting the stop-and-start playing that frequently occurs during sight-reading.
Conclusions
This study provides evidence that sub-optimal, experiential sight-reading practice has little impact on improvement of sight-reading skills in advanced pianists. Instead, deliberate practice approaches that facilitate practical connections between pitch and rhythm processing – such as playing by ear, improvisation, composition, and structural analyses – together with opportunities for collaborative playing, should be considered as alternatives for informal sight-reading practice.
Improving sight-reading skills is an important goal of music training, as the proficiency in this area has many beneficial flow-on effects, including the ability to learn new repertoire quickly, read complex scores (solo, chamber music, orchestral and choral scores), analyse musical structures and free up time typically spent on learning of the basic notes for more interesting collaborative and creative music activities. Therefore, a greater emphasis on developing this skill during undergraduate study is urgently needed across the globe to enhance student preparation for and integration into the music profession. The 21st century music graduates are now required to possess wide-ranging skills in order to fulfil multi-faceted roles in the globally-connected music industry (Burnard, 2014). Despite this rapidly changing environment, higher education training of classical musicians remains stubbornly focused on technique and solo repertoire (Bartleet et al., 2012; Gaunt, 2010). Instead, higher education institutions should strive to develop generic music skills such as sight-reading and related skills of improvisation, transposition, harmonisation, orchestration, composition, and collaborative playing to make music graduates work-ready (Zhukov, 2016).
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.