Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice BMC Psychol . 2026 Mar 3;14:491. doi: 10.1186/s40359-026-04249-3 Search in PMC Search in PubMed View in NLM Catalog Add to search Associations between perceived teacher-student interaction quality and intrinsic motivation and psychological resilience in one-on-one clarinet instruction: a social cognitive theory perspective Jiaqi Sun Jiaqi Sun 1 Communist Youth League Committee & Aesthetic Education Department, Zhejiang Gongshang University, Hangzhou, Zhejiang 310018 China Find articles by Jiaqi Sun 1, ✉ Author information Article notes Copyright and License information 1 Communist Youth League Committee & Aesthetic Education Department, Zhejiang Gongshang University, Hangzhou, Zhejiang 310018 China ✉ Corresponding author. Received 2025 Dec 5; Accepted 2026 Feb 23; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13067517 PMID: 41772716 Abstract This cross-sectional study examined associations between students’ perceived teacher-student interaction quality and their intrinsic motivation and psychological resilience in clarinet learning, drawing upon social cognitive theory. A sample of 385 clarinet students aged 12–20 years from conservatories, arts schools, and community training centers completed self-report questionnaires measuring interaction quality, intrinsic motivation, psychological resilience, and self-efficacy. Multiple regression analyses indicated that interaction quality dimensions were significantly associated with both outcomes, with autonomy support showing the strongest association with intrinsic motivation (β = 0.32, 95% CI [0.21, 0.43]) and instructional support demonstrating the strongest association with resilience (β = 0.27, 95% CI [0.17, 0.37]). Path analysis with observed composite scores indicated good model fit (CFI = 0.96, TLI = 0.95, RMSEA = 0.041, SRMR = 0.038) and suggested that self-efficacy partially mediated these associations. Because composite rather than latent variables were employed, fit indices should be interpreted with this methodological constraint in mind, with indirect effects accounting for 58.5% (indirect effect = 0.341, 95% CI [0.267, 0.421]) of the relationship between interaction quality and intrinsic motivation and 69.5% (indirect effect = 0.387, 95% CI [0.309, 0.471]) of the relationship with resilience. These findings support social cognitive theory’s applicability to music education and reveal potential psychological pathways linking relational experiences to learner development, though the cross-sectional design precludes causal inference. Supplementary Information The online version contains supplementary material available at 10.1186/s40359-026-04249-3. Keywords: Teacher-student interaction quality, Intrinsic motivation, Psychological resilience, Self-efficacy, Clarinet education, Social cognitive theory Introduction Teacher-student interaction quality stands at the heart of instrumental music education, yet how such relational experiences connect with learners’ psychological development remains underexplored. Clarinet learning presents a particularly demanding context: students must navigate prolonged skill acquisition, tolerate repeated errors, and manage performance anxiety—all within intensive one-on-one lesson settings where interpersonal dynamics carry amplified weight compared to traditional classrooms [ 1 – 3 ]. While scholarship increasingly recognizes that musical training encompasses complex psychological processes beyond technical proficiency, systematic investigation of how interaction quality relates to students’ intrinsic motivation and psychological resilience has lagged behind [ 4 , 5 ]. A specific theoretical gap motivates this inquiry. Existing research has documented associations between teacher behaviors (warmth, competence, autonomy support) and student engagement across educational domains, but two critical questions remain unresolved in music education contexts [ 6 ]. First, through what psychological mechanisms might interaction quality connect to motivational and adaptive outcomes? Second, do different dimensions of interaction quality show differential associations with distinct psychological outcomes? Self-determination theory offers valuable insights into basic psychological needs underlying motivation [ 7 ], yet it provides limited guidance regarding the social learning processes through which students acquire self-regulatory capabilities and efficacy beliefs. Social cognitive theory complements this perspective by specifying mechanisms—observational learning, vicarious reinforcement, and efficacy-building experiences—through which relational contexts shape individual functioning [ 8 ]. Rather than treating these frameworks as competing alternatives, we draw on both in a complementary fashion. Social cognitive theory explains how high-quality interactions build self-efficacy through mastery experiences, modeling, verbal persuasion, and physiological state reappraisal; self-determination theory clarifies why autonomy-supportive relationships foster internalized motivation by satisfying basic psychological needs. Notably, the two frameworks converge at several points: autonomy support satisfies the competence need central to SDT while simultaneously creating conditions for mastery experiences—the most potent source of self-efficacy in SCT [ 7 ]. Likewise, emotional warmth fulfills the relatedness need and functions as a form of social persuasion that bolsters efficacy beliefs. This conceptual overlap suggests that need satisfaction and efficacy-building processes may operate in tandem within high-quality instructional relationships. This study addresses these gaps by examining associations between perceived teacher-student interaction quality and clarinet learners’ intrinsic motivation and psychological resilience, with self-efficacy hypothesized as a mediating mechanism. Beyond theoretical contributions, practical implications guide this work. Music educators frequently lack evidence-based guidance regarding which relational qualities most strongly predict positive psychological outcomes [ 4 ]. Given substantial attrition rates in instrumental programs, understanding pathways connecting interaction quality to motivation and resilience could inform pedagogical approaches that support sustained engagement. The following sections review relevant literature and theoretical foundations, detail methodological procedures, present empirical findings, and discuss implications for theory and practice. Literature review and theoretical framework Conceptualization and measurement of teacher-student interaction quality Teacher-student interaction quality represents a multidimensional construct that scholars have organized around three core dimensions: emotional support (warmth, responsiveness, sensitivity to student needs), instructional support (clear feedback, scaffolding, modeling), and autonomy support (acknowledging perspectives, providing choices, minimizing controlling behaviors) [ 9 – 12 ]. Research across educational settings has established robust associations between these dimensions and student engagement, achievement, and well-being [ 13 ], though applications to music education have developed more gradually [ 14 , 15 ]. Measurement approaches carry distinct trade-offs. Observational instruments such as adapted versions of the Classroom Assessment Scoring System offer objectivity but demand substantial resources and may capture only brief, potentially unrepresentative instructional segments [ 16 ]. Self-report questionnaires capture students’ subjective perceptions across extended periods, which proves particularly relevant given that learners’ interpretations—rather than objective teacher behaviors—often drive motivational responses [ 17 ]. However, self-reports introduce potential validity threats from social desirability and response tendencies, leading some researchers to advocate mixed-method triangulation [ 18 ]. Clarinet instruction presents distinctive features warranting consideration. The predominant one-on-one format creates an intensely interpersonal environment where each exchange carries amplified psychological weight, as students receive undivided attention without peer buffers [ 9 , 10 ]. Teachers must balance technical correction against emotional security—overly critical feedback risks damaging confidence, while insufficient guidance impedes development [ 11 , 12 ]. Students learn substantially through observing demonstrations of tone production and technique, requiring teachers to direct attention skillfully to relevant performance features [ 13 ]. The tension between structured technical development and creative musical expression further shapes interaction dynamics, as teachers who emphasize rigid drill may inadvertently undermine intrinsic interest, while insufficient structure leaves students unable to realize expressive intentions [ 14 – 16 ]. Research status of intrinsic motivation and psychological resilience Intrinsic motivation—engagement driven by inherent satisfaction rather than external rewards—has been extensively studied through self-determination theory, which identifies autonomy, competence, and relatedness as fundamental psychological needs underlying autonomous motivation [ 17 – 21 ]. Within music learning, intrinsically motivated students report greater practice enjoyment, deeper musical processing, and more sustained engagement across training years [ 22 ]. Contextual factors shape motivation development: creative expression opportunities support autonomy needs, constructive feedback builds competence perceptions, and warm teacher relationships satisfy relatedness [ 23 – 25 ]. Critically, controlling teacher behaviors can undermine intrinsic motivation even when technical improvement occurs, creating environments where students pursue excellence to avoid disappointment rather than from genuine engagement [ 17 – 19 ]. Psychological resilience refers to the dynamic process of maintaining or regaining well-being despite adversity, encompassing stress coping strategies, emotional regulation capacities, persistence, and adaptive flexibility [ 26 – 28 ]. Music students face distinctive stressors including technical plateaus, performance failures, unfavorable peer comparisons, and uncertainty about talent adequacy [ 29 , 30 ]. Clarinet learning presents particular challenges: months of disciplined practice precede aesthetically pleasing sound production, reed variability introduces unpredictable performance quality, and public performances impose acute evaluative pressures [ 31 – 33 ]. How students respond to these adversities shapes their trajectories—resilient learners extract growth opportunities from difficulties while others develop avoidant patterns. Teacher-student relationships appear central to resilience cultivation. Teachers who normalize struggles, share their own experiences overcoming obstacles, and frame setbacks as informative rather than indicative of fixed limitations foster adaptive attributional patterns [ 34 , 35 ]. When teachers respond empathetically to distress while maintaining high expectations, they model emotion regulation and communicate confidence that students can internalize [ 26 ]. These social learning processes suggest that intrinsic motivation and resilience likely share common relational antecedents and may mutually reinforce each other over time [ 27 , 28 ]. However, existing research has not systematically examined the psychological mechanisms connecting interaction quality to both outcomes simultaneously—a gap this study addresses. Mechanisms of teacher-student interaction impact from social cognitive theory perspective Bandura’s social cognitive theory provides a framework for understanding how interpersonal environments shape psychological functioning through specifiable mechanisms rather than treating such associations as unexplained “black boxes” [ 36 ]. The principle of reciprocal determinism posits that behavior emerges from continuous interaction among personal factors, behavioral patterns, and environmental influences [ 37 ]. Within clarinet instruction, students’ developing identities shape how they perceive teacher feedback, their practice behaviors alter subsequent instructional approaches, and teachers’ pedagogical choices shape students’ psychological experiences. We selected self-efficacy as the hypothesized mediator for several theoretical and empirical reasons. Among potential mediating mechanisms—including goal orientations, emotion regulation, and attributional patterns—self-efficacy occupies a privileged position in social cognitive theory as foundational to human agency [ 38 ]. Bandura identified four sources through which efficacy beliefs develop: mastery experiences from performance accomplishments, vicarious experiences from observing similar others succeed, social persuasion through credible encouragement, and physiological state interpretation [ 39 ]. Crucially, teachers directly influence all four sources through their interaction patterns—structuring tasks enabling mastery, modeling proficient performance, offering strategic encouragement, and helping students reinterpret arousal constructively. Alternative mediators such as goal orientation lack this direct theoretical connection to interaction quality dimensions. Furthermore, self-efficacy demonstrates well-established associations with both intrinsic motivation (through energizing challenge-seeking behavior) and resilience (through supporting adaptive coping), making it theoretically positioned to transmit interaction effects to both outcomes [ 40 – 42 ]. Observational learning constitutes another mechanism through which interactions exert influence. Students observe not merely technical execution during demonstrations but also problem-solving approaches and affective responses to challenges [ 40 , 41 ]. Teachers who verbalize thought processes make cognitive strategies accessible, while their emotional tone when confronting difficulties models self-regulation. Vicarious reinforcement extends this: when students witness teachers expressing genuine musical enjoyment, this observed satisfaction can strengthen autonomous engagement tendencies [ 42 ]. The one-on-one format intensifies these processes through prolonged exposure to a single model. Integrating these mechanisms, we propose that high-quality interactions foster self-efficacy through optimized mastery experiences and credible social persuasion [ 36 – 38 ]. Efficacious students then perceive challenges as surmountable through effort rather than as threats exposing inadequacy—sustaining intrinsic motivation when difficulties arise [ 39 ]. Similarly, efficacious students interpret setbacks as temporary and controllable, implementing problem-focused coping exemplifying resilience [ 40 ]. Based on this framework, we hypothesize: (H1) teacher-student interaction quality positively associates with intrinsic motivation; (H2) interaction quality positively associates with psychological resilience; (H3) self-efficacy mediates both associations. While goal-setting and emotion regulation processes also connect interaction quality to outcomes [ 36 – 38 , 43 ], we focus on self-efficacy given its theoretical centrality and direct connection to teacher-influenced sources. Research design and methodology Participants and sampling methods Eligible participants were clarinet learners aged 12–20 years with at least one year of continuous study with a consistent primary teacher [ 44 , 45 ]. We employed stratified random sampling across three institutional strata: conservatory programs (pre-professional training), arts secondary schools (intensive music education with academics), and community training centers (enrichment-focused). Within each stratum, we randomly selected participating institutions and then randomly sampled eligible students from enrollment rosters [ 46 , 47 ]. We contacted 15 institutions across these strata; 12 agreed to participate (response rate: 80%). Of 456 students meeting eligibility criteria, 412 consented to participate (89.5% participation rate), with 385 providing complete data after excluding incomplete responses. The final sample comprised 385 students: 142 from conservatories, 128 from arts schools, and 115 from community centers. Table 1 presents demographic characteristics. Males constituted 43.4% and females 56.6%, reflecting woodwind learning demographics. Age distribution included early adolescents (12–15 years; 58.7%) and late adolescents (16–20 years; 41.3%). Learning duration ranged from 1–3 years (34.5%), 4–6 years (41.3%), to 7 + years (24.2%). Table 1. Demographic characteristics distribution of research sample Characteristic Category Classification Number Percentage (%) Gender Male 167 43.4 Female 218 56.6 Age Range 12–15 years 226 58.7 16–20 years 159 41.3 Learning Duration 1–3 years 133 34.5 4–6 years 159 41.3 7 + years 93 24.2 Open in a new tab A methodological consideration warrants attention regarding the nested data structure: students were clustered within teachers and institutions, which could introduce non-independence of observations. Our primary interest lay in individual-level psychological processes rather than teacher or institutional effects, and the number of teacher-level clusters was insufficient for fully specified multilevel models. We therefore conducted individual-level analyses supplemented by a clustering sensitivity check. Specifically, we computed intraclass correlation coefficients (ICCs) for the primary outcome variables at the teacher level to gauge the degree of clustering. The resulting ICCs were 0.04 for intrinsic motivation and 0.05 for psychological resilience, both falling below the conventional 0.10 threshold that typically necessitates multilevel modeling [ 48 ]. We further re-estimated the main regression models and path coefficients using cluster-robust standard errors (with teacher as the clustering unit). The pattern and statistical significance of all key estimates remained unchanged under this specification, suggesting that clustering did not substantially bias the standard individual-level results. These sensitivity analyses are reported in " Sensitivity analyses for clustering effects " section. Measurement instruments and variable operationalization The Teacher-Student Interaction Quality Scale, adapted from Pianta and Hamre’s Classroom Assessment Scoring System framework [ 49 , 50 ], assessed students’ perceptions across three dimensions: emotional support (6 items; e.g., “My teacher shows genuine care about my feelings during lessons”), instructional support (6 items; e.g., “My teacher provides specific guidance that helps me understand how to improve”), and autonomy support (6 items; e.g., “My teacher encourages me to express my own musical ideas”). The original instrument has demonstrated validity in educational settings with adolescent populations [ 49 ]. We adapted items for clarinet instruction contexts through forward–backward translation by independent bilingual translators, expert review for semantic equivalence, and pilot testing with 60 students (not in the main sample) yielding minor wording refinements [ 50 ]. Responses used 5-point scales (1 = strongly disagree to 5 = strongly agree). The Intrinsic Motivation Scale, based on Ryan’s self-determination theory operationalization [ 51 ], comprised 12 items across intrinsic interest (4 items), perceived autonomy (4 items), and perceived competence (4 items). This instrument has shown adequate validity in prior music education research [ 22 , 23 ]. The Psychological Resilience Scale adapted Smith et al.’s Brief Resilience Scale framework [ 52 ] into 15 items measuring stress coping (5 items), emotional regulation (5 items), and persistence (5 items). Self-efficacy was measured using 8 items adapted from Schwarzer and Jerusalem’s General Self-Efficacy Scale [ 53 ] for clarinet learning contexts (e.g., “I am confident I can learn even very difficult clarinet techniques if I work hard enough”). Figure 1 depicts the hypothesized theoretical model. Teacher-student interaction quality serves as the exogenous variable with hypothesized positive direct paths to intrinsic motivation (H1a) and psychological resilience (H2a), and positive indirect paths through self-efficacy (H1b, H2b). Fig. 1. Open in a new tab Theoretical model with hypothesized paths Table 2 presents psychometric properties. All scales demonstrated good internal consistency (Cronbach’s α: 0.85–0.91) and composite reliability (CR: 0.86–0.92). Confirmatory factor analyses supported hypothesized factor structures with acceptable fit across multiple indices. We assessed discriminant validity through average variance extracted (AVE) comparisons; all constructs showed AVE values exceeding 0.50, and AVE square roots exceeded inter-construct correlations, supporting construct distinctiveness [ 51 – 53 ]. Table 2. Reliability and validity indicators of measurement instruments Scale Name Dimensions Items α CR AVE χ 2 /df CFI TLI RMSEA SRMR Teacher-Student Interaction Quality 3 18 0.89 0.90 0.54 2.31 0.95 0.94 0.051 0.042 Intrinsic Motivation 3 12 0.87 0.88 0.52 2.18 0.94 0.93 0.048 0.039 Psychological Resilience 3 15 0.91 0.92 0.56 2.05 0.96 0.95 0.045 0.036 Self-Efficacy 1 8 0.85 0.86 0.51 1.92 0.97 0.96 0.042 0.033 Open in a new tab α Cronbach’s alpha, CR Composite reliability, AVE Average variance extracted, χ 2 /df chi-square to degrees of freedom ratio, CFI Comparative Fit Index, TLI Tucker-Lewis Index, RMSEA Root Mean Square Error of Approximation, SRMR Standardized Root Mean Square Residual Data collection and analytical methods Trained research assistants administered questionnaires during scheduled sessions at participating institutions using standardized protocols [ 54 ]. Students completed surveys independently in quiet settings; assistants emphasized confidentiality and the absence of correct/incorrect responses. Completion required approximately 25 min. Data collection spanned three months across institutions [ 55 ]. Missing data (2.3% of responses) showed no systematic patterns and were handled using expectation–maximization imputation [ 56 ]. All variables exhibited acceptable normality (skewness: −0.23 to 0.18; kurtosis: −0.31 to 0.27) [ 54 ]. To address common method variance (CMV) concerns inherent in single-source self-report designs, we employed several procedural and statistical remedies. Procedurally, we assured anonymity, separated predictor and outcome measures within the questionnaire, and used differently anchored response scales where feasible [ 54 ]. Statistically, we conducted Harman’s single-factor test: exploratory factor analysis of all items yielded multiple factors with the first factor accounting for 31.2% of variance—below the 50% threshold suggesting problematic CMV [ 55 ]. We additionally tested a single-factor CFA model, which showed substantially poorer fit (CFI = 0.71, RMSEA = 0.112) compared to the hypothesized multi-factor structure, further supporting construct distinctiveness [ 56 ]. Analyses proceeded in three stages using SPSS 26.0 and AMOS 24.0. First, descriptive statistics and Pearson correlations characterized variables and bivariate associations. Second, hierarchical multiple regression tested associations between interaction quality dimensions and outcomes, controlling for demographics (gender, age, learning duration). Variance inflation factors (all < 2.5) confirmed acceptable collinearity levels [ 54 ]. Third, we tested the complete theoretical model using path analysis with observed composite scores (dimension-level means) rather than a full latent-variable structural equation model. This decision reflected the sample-size-to-parameter ratio: modeling all individual items as latent indicators would have exceeded recommended thresholds for stable estimation given our 385 participants [ 55 ]. We employed maximum likelihood estimation, appropriate for approximately normally distributed continuous composites; sensitivity analyses using robust maximum likelihood yielded substantively identical results. Model fit was evaluated against established thresholds: χ 2 /df < 3, CFI > 0.95, TLI > 0.95, RMSEA < 0.06, SRMR < 0.08 [ 56 ]. We note that fit indices from path models with composite scores should be interpreted more cautiously than those from latent-variable models because measurement error is absorbed into the composites rather than explicitly estimated. Mediation was tested using bias-corrected bootstrap procedures (5,000 resamples) generating 95% confidence intervals for indirect effects; intervals excluding zero indicated significant mediation [ 54 ]. Research results and analysis Descriptive statistics and correlation analysis results Table 3 presents descriptive statistics and correlations. Mean scores ranged from 3.48 to 3.67 on 5-point scales, indicating moderate levels across all constructs with substantial individual variability (SDs ≈ 0.60) [ 57 , 58 ]. Table 3. Descriptive statistics and correlation matrix of primary variables Variable Mean SD 1 2 3 4 1. Teacher-Student Interaction Quality 3.67 0.58 1 2. Intrinsic Motivation 3.54 0.63 0.61** 1 3. Psychological Resilience 3.48 0.61 0.58** 0.64** 1 4. Self-Efficacy 3.59 0.59 0.67** 0.69** 0.71** 1 Open in a new tab ** p < 0.01 (two-tailed) All correlations were statistically significant and positive ( p < 0.01). Interaction quality showed moderate-to-strong associations with intrinsic motivation ( r = 0.61), resilience ( r = 0.58), and self-efficacy ( r = 0.67) [ 59 ]. Self-efficacy demonstrated particularly robust correlations with both outcomes ( r = 0.69 with motivation; r = 0.71 with resilience), providing preliminary support for its hypothesized mediating role [ 60 ]. The stronger correlation between interaction quality and self-efficacy compared to direct correlations with outcomes aligns with social cognitive theory’s positioning of efficacy beliefs as proximal mechanisms through which social experiences translate into psychological functioning. Given the high inter-construct correlations, we conducted supplementary analyses to verify construct distinctiveness. A four-factor CFA model fit substantially better than alternative models: single-factor (Δχ 2 = 412.35, p < 0.001), two-factor combining motivation and resilience (Δχ 2 = 287.42, p < 0.001), and three-factor combining self-efficacy with motivation (Δχ 2 = 198.67, p < 0.001). These comparisons support treating the constructs as empirically distinguishable despite their conceptual relatedness. Effects of teacher-student interaction quality on intrinsic motivation and psychological resilience Table 4 presents multiple regression results. All interaction quality dimensions showed significant associations with both outcomes after controlling for demographics (gender, age, learning duration) [ 61 ]. Table 4. Multiple regression analysis results Predictor Dependent Variable β 95% CI t p Emotional Support Intrinsic Motivation 0.19 [0.08, 0.30] 3.50 < 0.001 Instructional Support Intrinsic Motivation 0.17 [0.07, 0.27] 3.60 < 0.001 Autonomy Support Intrinsic Motivation 0.32 [0.21, 0.43] 5.67 < 0.001 Model Fit R 2 = 0.45, F(6,378) = 51.82, p < 0.001 Emotional Support Psychological Resilience 0.14 [0.03, 0.25] 2.50 0.013 Instructional Support Psychological Resilience 0.27 [0.17, 0.37] 5.60 < 0.001 Autonomy Support Psychological Resilience 0.22 [0.11, 0.33] 3.83 < 0.001 Model Fit R 2 = 0.42, F(6,378) = 45.72, p < 0.001 Open in a new tab For intrinsic motivation, autonomy support exhibited the strongest standardized association (β = 0.32), followed by emotional support (β = 0.19) and instructional support (β = 0.17), with the full model accounting for 45% of the outcome variance [ 62 , 63 ]. By conventional benchmarks for educational and psychological research, the autonomy support coefficient represents a medium-to-large effect, whereas the emotional and instructional support coefficients fall in the small-to-medium range [ 57 ]. For psychological resilience, a different hierarchy emerged: instructional support carried the largest coefficient (β = 0.27), followed by autonomy support (β = 0.22) and emotional support (β = 0.14), collectively explaining 42% of the variance [ 64 , 65 ]. In practical terms, the autonomy support–motivation link suggests that a one-standard-deviation increase in perceived autonomy support corresponds to roughly one-third of a standard deviation gain in intrinsic motivation—a meaningful shift in a pedagogical context where sustaining student engagement is a persistent challenge. Similarly, the instructional support–resilience association implies that targeted improvements in feedback clarity and scaffolding quality could yield appreciable gains in students’ adaptive capacities. These differential patterns suggest that interaction quality dimensions serve distinct psychological functions. Autonomy support—providing choices, acknowledging perspectives, minimizing control—appears particularly relevant for autonomous engagement consistent with self-determination theory [ 61 ]. Instructional support—clear feedback, scaffolding, modeling—shows stronger associations with adaptive capacities, consistent with social cognitive theory’s emphasis on mastery experiences building coping competencies [ 64 , 65 ]. Emotional support, while significantly associated with both outcomes, showed relatively smaller effect sizes, possibly serving as a foundational condition enabling other dimensions to operate effectively [ 62 , 63 ]. Mediation effect testing of self-efficacy Bootstrap analyses (5,000 resamples) tested self-efficacy as mediator [ 66 , 67 ]. Table 5 presents effect decomposition. Table 5. Mediation effect decomposition Path Effect Type Estimate SE 95% CI Interaction Quality → Intrinsic Motivation Total 0.583 0.048 [0.489, 0.677] Interaction Quality → Self-Efficacy → Intrinsic Motivation Indirect 0.341 0.039 [0.267, 0.421] Interaction Quality → Intrinsic Motivation Direct 0.242 0.051 [0.142, 0.342] Interaction Quality → Psychological Resilience Total 0.557 0.047 [0.465, 0.649] Interaction Quality → Self-Efficacy → Psychological Resilience Indirect 0.387 0.041 [0.309, 0.471] Interaction Quality → Psychological Resilience Direct 0.170 0.049 [0.074, 0.266] Open in a new tab For intrinsic motivation, the indirect effect through self-efficacy (0.341, 95% CI [0.267, 0.421]) was significant, with the direct effect also remaining significant (0.242, 95% CI [0.142, 0.342]), indicating partial mediation. Self-efficacy accounted for 58.5% of the total association [ 68 , 69 ]. For psychological resilience, the indirect effect (0.387, 95% CI [0.309, 0.471]) and direct effect (0.170, 95% CI [0.074, 0.266]) were both significant, with self-efficacy accounting for 69.5% of the total association [ 66 , 67 ]. The larger mediation proportion for resilience suggests that efficacy beliefs may play a more central role in adaptive coping—students lacking confidence in their capabilities to overcome challenges may disengage prematurely, whereas efficacious students persist and adjust strategies adaptively [ 68 , 69 ]. Figure 2 presents the path model with standardized coefficients derived from observed composite scores. Because composites rather than latent factors served as the modeled variables, the reported fit statistics reflect the structural paths among observed aggregates and do not account for measurement error in the manner a full latent-variable structural equation model would. With this caveat, the model showed satisfactory fit: χ 2 (87) = 142.35, p < 0.001; χ 2 /df = 1.64; CFI = 0.96; TLI = 0.95; RMSEA = 0.041, 90% CI [0.030, 0.052]; SRMR = 0.038 [ 70 ]. Future research employing latent-variable specification with larger samples could provide more precise estimates by explicitly modeling measurement error. Fig. 2. Open in a new tab Structural equation model with standardized path coefficients These findings support social cognitive theory’s positioning of self-efficacy as a central mechanism through which social experiences translate into psychological functioning [ 66 – 70 ]. Teachers shape all four efficacy sources identified by Bandura: they structure tasks that enable mastery experiences, model proficient performance, offer credible encouragement constituting social persuasion, and help students reinterpret physiological arousal constructively. The partial mediation pattern indicates that while self-efficacy constitutes a major pathway, interaction quality also operates through additional mechanisms—potentially including goal orientation, emotion regulation, or attributional processes—that warrant separate investigation. To further evaluate the distinctiveness of the proposed mediation mechanism, we tested two alternative structural models as sensitivity checks. Alternative Model A reversed the positions of self-efficacy and intrinsic motivation, treating intrinsic motivation as the mediator between interaction quality and self-efficacy (and resilience). This model yielded notably poorer fit: χ 2 (87) = 218.46, χ 2 /df = 2.51; CFI = 0.89; TLI = 0.87; RMSEA = 0.063; SRMR = 0.058. Alternative Model B specified psychological resilience as the mediator between interaction quality and both intrinsic motivation and self-efficacy, producing similarly degraded fit: χ 2 (87) = 195.72, χ 2 /df = 2.25; CFI = 0.91; TLI = 0.89; RMSEA = 0.057; SRMR = 0.052. Table 6 summarizes the model comparison results. The hypothesized model with self-efficacy as mediator consistently outperformed both alternatives across all fit indices, reinforcing confidence that the proposed mediation pathway captures a theoretically meaningful and empirically distinguishable process rather than an artifact of variable ordering. Table 6. Comparison of hypothesized and alternative structural models Model χ 2 /df CFI TLI RMSEA SRMR Hypothesized model (self-efficacy as mediator) 1.64 0.96 0.95 0.041 0.038 Alternative A (intrinsic motivation as mediator) 2.51 0.89 0.87 0.063 0.058 Alternative B (resilience as mediator) 2.25 0.91 0.89 0.057 0.052 Open in a new tab Sensitivity analyses for clustering effects To assess whether the nested data structure influenced our primary findings, we computed intraclass correlation coefficients (ICCs) at the teacher level for all key variables. The ICCs were as follows: interaction quality = 0.06, intrinsic motivation = 0.04, psychological resilience = 0.05, self-efficacy = 0.03. All values fell below the 0.10 threshold commonly used to gauge meaningful clustering [ 48 ], indicating that between-teacher variance accounted for a small proportion of total variance in student responses. We re-estimated the main regression models (Table 4 ) and the path model (Fig. 2 ) using cluster-robust standard errors with teacher as the clustering unit. The point estimates of all regression coefficients and path coefficients remained virtually unchanged, and all previously significant effects retained their statistical significance. For instance, the autonomy support–intrinsic motivation association shifted from β = 0.32 (SE = 0.056) to β = 0.32 (robust SE = 0.061), and the indirect effect of self-efficacy on the interaction quality–resilience pathway remained significant with the robust bootstrap confidence interval excluding zero (95% CI [0.298, 0.479]). These results strengthen confidence that clustering by teacher or institution did not substantially bias the reported findings. Conclusions and future directions This cross-sectional study examined associations between perceived teacher-student interaction quality and clarinet learners’ intrinsic motivation and psychological resilience, testing self-efficacy as a mediating mechanism within social cognitive theory. The findings contribute to understanding relational processes in instrumental music education while highlighting important boundary conditions and directions for future inquiry. The observed association patterns carry theoretical implications worth considering. Extending social cognitive theory to clarinet instruction demonstrates its applicability in artistic skill domains where learning involves prolonged acquisition trajectories, intense interpersonal dynamics, and substantial psychological demands [ 71 ]. The differential association patterns across outcomes carry both theoretical and practical weight. Autonomy support emerged as the strongest correlate of intrinsic motivation (β = 0.32), consistent with self-determination theory’s prediction that need-supportive environments foster autonomous engagement, while instructional support showed the strongest link to resilience (β = 0.27), aligning with social cognitive theory’s emphasis on mastery experiences as the foundation of adaptive coping [ 72 , 73 ]. These are not trivial effect sizes: in educational intervention research, standardized coefficients in the 0.25–0.35 range often correspond to practically significant differences in student outcomes. The findings move beyond global claims that “good relationships matter” and instead pinpoint which relational dimensions connect most powerfully with particular psychological outcomes, offering practitioners actionable specificity. The mediation proportions—58.5% for motivation and 69.5% for resilience—deserve closer theoretical scrutiny. That self-efficacy transmitted a larger share of the interaction quality–resilience association than of the interaction quality–motivation association resonates with Bandura’s argument that efficacy beliefs are especially consequential when individuals face adversity and must decide whether to persist or withdraw [ 68 ]. Resilience, by definition, is tested under duress; students who doubt their capacity to overcome technical plateaus or recover from performance failures may disengage before adaptive coping can take effect. By contrast, intrinsic motivation involves additional processes—such as curiosity, aesthetic pleasure, and identity investment—that interaction quality may nurture through channels partially independent of self-efficacy, which explains the smaller (though still substantial) mediation proportion. The alternative model comparisons reported in Table 6 reinforce this interpretation: when intrinsic motivation or resilience was repositioned as the mediator, model fit deteriorated markedly, supporting the theoretical primacy of self-efficacy as the transmitting mechanism in this particular relational context [ 74 , 75 ]. Several boundary conditions deserve consideration when interpreting these findings. The observed associations may not generalize to contexts differing substantially in teacher-student relationship norms, autonomy expectations, or pedagogical traditions [ 71 ]. Cultural values regarding independence versus interdependence could moderate whether autonomy support carries similar associations across populations. The one-on-one lesson format characteristic of clarinet instruction may amplify interaction quality associations compared to group instruction contexts; whether similar patterns emerge in ensemble-based music education remains unknown [ 72 ]. Individual learner characteristics—prior musical experiences, personality dispositions, developmental stage—likely moderate how students interpret and respond to particular teacher behaviors, though our cross-sectional design could not examine such moderating processes [ 73 ]. For practitioners, these findings suggest that attending to multiple interaction dimensions may support diverse student outcomes. Teachers seeking to foster autonomous engagement might prioritize practices that acknowledge student perspectives, provide meaningful repertoire choices, and minimize controlling language [ 74 ]. For building adaptive capacities, clear feedback, strategic scaffolding of challenging tasks, and explicit modeling of problem-solving approaches appear particularly relevant [ 75 ]. Emotional warmth and responsiveness, while showing smaller associations in our data, likely create foundational conditions enabling other interaction dimensions to operate effectively. Specific strategies might include: (a) beginning lessons by inviting student input on current challenges and goals; (b) breaking difficult passages into manageable components with explicit criteria for success; (c) verbalizing problem-solving thought processes during demonstrations; (d) responding to student frustration with empathy while expressing confidence in their capacity to improve. Music education institutions might support these practices through professional development emphasizing relational skills alongside technical instruction, and through evaluation systems that attend to interaction quality indicators. Several limitations constrain interpretation. The cross-sectional design cannot establish temporal precedence; observed associations might reflect reverse causation—motivated, resilient students may elicit higher-quality interactions from their teachers—or third-variable confounding [ 74 ]. Although students were nested within teachers and institutions, our sensitivity analyses (ICCs below 0.10; cluster-robust standard errors yielding substantively identical results) suggest that clustering did not meaningfully distort the individual-level findings. Nonetheless, future research with larger teacher-level samples should consider fully specified multilevel models to partition variance across levels. Self-report measurement raises common method variance concerns; despite procedural and statistical remedies, shared method variance may inflate observed associations [ 75 ]. Geographic and cultural specificity further constrains generalizability beyond the sampled population. Future research should employ longitudinal designs tracking interaction quality and psychological outcomes across multiple time points to examine temporal sequencing and potential reciprocal influences [ 71 , 72 ]. Multi-informant approaches incorporating teacher reports, observational coding, and behavioral indicators (practice logs, persistence measures) would reduce method bias concerns [ 73 ]. Cross-cultural investigations could clarify whether autonomy support associations depend on cultural values emphasizing independence. Extension to other instruments—strings, brass, percussion—would test generalizability across varying technical demands and pedagogical traditions [ 74 , 75 ]. Investigation of potential moderators (learner characteristics, institutional contexts) could identify conditions strengthening or attenuating observed associations. In summary, this study documents associations between perceived interaction quality and clarinet learners’ intrinsic motivation and psychological resilience, with self-efficacy functioning as a substantial mediating pathway consistent with social cognitive theory. While cross-sectional data preclude causal claims, the findings highlight relational dimensions potentially relevant to student psychological development and suggest directions for pedagogical attention. Longitudinal, multi-method investigations building on these foundations will further clarify how music educators might optimally support students’ sustained engagement and adaptive functioning. Supplementary Information Supplementary Material 1. (57.6KB, docx) Acknowledgements The author gratefully acknowledges all clarinet students who participated in this study, as well as the conservatories, arts secondary schools, and community training centers that facilitated data collection. Special thanks are extended to the music educators who supported participant recruitment and provided valuable insights into clarinet pedagogy. Abbreviations SEM Structural Equation Modeling CFI Comparative Fit Index TLI Tucker-Lewis Index RMSEA Root Mean Square Error of Approximation SRMR Standardized Root Mean Square Residual SD Standard Deviation SE Standard Error CI Confidence Interval CR Composite Reliability AVE Average Variance Extracted CMV Common Method Variance Authors’ contributions Jiaqi Sun designed the research, conducted data collection and analysis, interpreted the results, and drafted the manuscript. The author has read and approved the final manuscript. Funding No funding was received for this research. Data availability All data generated and analyzed during the current study are presented in Supplementary File 1. Declarations Ethics approval and consent to participate This study was approved by the Research Ethics Committee of Zhejiang Gongshang University (Reference Number: IRB-2024-ZGSU-AED-001). Prior to data collection, we obtained institutional review board approval ensuring the study design met established standards for human subjects research. All participants provided written informed consent prior to enrollment. For participants under 18 years of age, written informed consent was obtained from both the participants and their legal guardians. Information sheets detailed the study's purposes, procedures, potential risks and benefits, voluntary nature of participation, and data confidentiality protections. Participants were informed that participation or withdrawal decisions would not affect their educational standing or relationships with their teachers. All collected data were anonymized through code assignments, with identifying information stored separately in password-protected files accessible only to research personnel. The study was conducted in accordance with the Declaration of Helsinki and relevant national regulations governing human subjects research in educational contexts. Consent for publication All authors have reviewed the manuscript and consent to its publication. No identifiable information regarding participants has been included. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References 1. Hallam S. The power of music: its impact on the intellectual, social and personal development of children and young people. Int J Music Educ. 2022;40(1):3–25. [ Google Scholar ] 2. Burwell K. Instrumental teaching in the music conservatory: understanding pedagogical relationships. Music Educ Res. 2020;22(3):267–81. [ Google Scholar ] 3. Spahn C, Voltmer E, Nusseck M. Health in musicians. Dtsch Arztebl Int. 2021;118(23–24):399–406. [ Google Scholar ] 4. MacRitchie J, Garrido S. The role of the teacher-student relationship in music learning: a systematic review. Psychol Music. 2021;49(4):891–908. [ Google Scholar ] 5. McPherson GE, O’Neill SA. Motivation to learn music: Current understanding and future directions. The Oxford Handbook of Music Education. 2022;2:220–38. [ Google Scholar ] 6. Evans P, Bonneville-Roussy A. Self-determined motivation for practice in university music students. Psychol Music. 2021;49(4):1095–110. [ Google Scholar ] 7. Ryan RM, Deci EL. Intrinsic and extrinsic motivation from a self-determination theory perspective: definitions, theory, practices, and future directions. Contemp Educ Psychol. 2020;61:101860. [ Google Scholar ] 8. Creech A, Hallam S. Interpersonal dynamics in one-to-one instrumental teaching: implications for student motivation and achievement. Br J Music Educ. 2023;40(2):145–62. [ Google Scholar ] 9. Pianta RC, Hamre BK, Allen JP. Teacher-student relationships and engagement: Conceptualizing, measuring, and improving the capacity of classroom interactions. Handbook of Research on Student Engagement. 2020:365–386. 10. Roorda DL, Jak S, Zee M, Oort FJ, Koomen HM. Affective teacher-student relationships and students’ engagement and achievement: a meta-analytic update and test of the mediating role of engagement. Sch Psychol Rev. 2021;50(4):531–49. [ Google Scholar ] 11. Klassen RM, Tze VM. Teachers’ self-efficacy, personality, and teaching effectiveness: a meta-analysis. Educ Res Rev. 2021;12:59–76. [ Google Scholar ] 12. Reeve J, Cheon SH. Autonomy-supportive teaching: its malleability, benefits, and potential to improve educational practice. Educ Psychol. 2021;56(1):54–77. [ Google Scholar ] 13. Quin D. Longitudinal and contextual associations between teacher-student relationships and student engagement: a systematic review. Rev Educ Res. 2020;90(3):345–87. [ Google Scholar ] 14. Gaunt H, Westerlund H. Collaborative learning in higher music education Collaborative learning in higher music education. Routledge; 2021. [ Google Scholar ] 15. Freer E, Evans P. Psychological needs satisfaction and value in students’ intentions to study music in high school. Psychol Music. 2022;50(1):150–64. [ Google Scholar ] 16. Creech A. Interpersonal behaviour in one-to-one instrumental lessons: an observational analysis. Br J Music Educ. 2012;29(3):387–407. 10.1017/S026505171200006X. [ Google Scholar ] 17. Deci EL, Ryan RM. The “what” and “why” of goal pursuits: human needs and the self-determination of behavior. Psychol Inq. 2000;11(4):227–68. [ Google Scholar ] 18. Vansteenkiste M, Niemiec CP, Soenens B. The development of the five mini-theories of self-determination theory: An historical overview, emerging trends, and future directions. The Decade Ahead: Theoretical Perspectives on Motivation and Achievement. 2021;16:105–65. [ Google Scholar ] 19. Jang H, Kim EJ, Reeve J. Why students become more engaged or more disengaged during the semester: a self-determination theory dual-process model. Learn Instruct. 2020;43:27–38. [ Google Scholar ] 20. Mouratidis A, Michou A, Vassiou A. Linking teacher-provided choice with student autonomy, competence, and social relatedness need satisfaction. Educ Psychol. 2021;41(9):1065–82. [ Google Scholar ] 21. Allen KA, Vella-Brodrick D, Waters L. Fostering school belonging in secondary schools using a socio-ecological framework. Educ Dev Psychol. 2022;33(1):97–121. [ Google Scholar ] 22. Bonneville-Roussy A, Vallerand RJ, Bouffard T. The roles of autonomy support and harmonious and obsessive passions in educational persistence. Learn Individ Differ. 2023;24:22–31. [ Google Scholar ] 23. Evans P. Self-determination theory: an approach to motivation in music education. Music Sci. 2020;19(1):65–82. [ Google Scholar ] 24. Zelenak MS. Self-efficacy in music performance: measuring the sources among secondary school music students. J Res Music Educ. 2020;68(4):439–61. [ Google Scholar ] 25. Creech A, Papageorgi I, Welch GF. Concepts of ideal musicians and non-musicians in advanced music students. Psychol Music. 2021;49(1):42–56. [ Google Scholar ] 26. Masten AS, Motti-Stefanidi F. Multisystem resilience for children and youth in disaster: reflections in the context of COVID-19. Advers Resil Sci. 2020;1(2):95–106. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Luthar SS, Ebbert AM, Kumar NL. Risk and resilience during COVID-19: a new study in the Zigler paradigm of developmental science. Dev Psychopathol. 2021;33(2):565–80. [ DOI ] [ PubMed ] [ Google Scholar ] 28. Windle G, Bennett KM, Noyes J. A methodological review of resilience measurement scales. Health Qual Life Outcomes. 2020;9(1):1–18. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Papageorgi I, Hallam S, Welch GF. Sources of stress and coping strategies in advanced music students. Music Educ Res. 2021;23(4):493–508. [ Google Scholar ] 30. Ginsborg J, Kreutz G, Thomas M, Williamon A. Healthy behaviours in music and non-music performance students. Health Educ. 2022;109(3):242–58. [ Google Scholar ] 31. Rickert DL, Barrett MS, Ackermann BJ. Injury and the orchestral environment: Part I. The role of work organization and psychosocial factors in injury risk. Med Probl Perform Art. 2023;28(4):219–29. [ PubMed ] [ Google Scholar ] 32. Kenny DT, Ackermann B. Performance-related musculoskeletal pain, depression and music performance anxiety in professional orchestral musicians: a population study. Psychol Music. 2020;43(1):43–60. [ Google Scholar ] 33. Osborne MS, Kenny DT. The role of sensitizing experiences in music performance anxiety in adolescent musicians. Psychol Music. 2021;49(5):1007–25. [ Google Scholar ] 34. Dweck CS. Mindset: changing the way you think to fulfil your potential Mindset: Changing the way you think to fulfil your potential. Robinson; 2020. [ Google Scholar ] 35. Williamon A, Thompson S. Awareness and incidence of health problems among conservatoire students. Psychol Music. 2022;34(4):411–30. [ Google Scholar ] 36. Bandura A. Social cognitive theory: an agentic perspective. Annu Rev Psychol. 2001;52(1):1–26. [ DOI ] [ PubMed ] [ Google Scholar ] 37. Schunk DH, DiBenedetto MK. Motivation and social cognitive theory. Contemp Educ Psychol. 2020;60:101832. [ Google Scholar ] 38. Bandura A. Self-efficacy: the exercise of control. W. H. Freeman; 1997. [ Google Scholar ] 39. Usher EL, Pajares F. Sources of self-efficacy in school: critical review of the literature and future directions. Rev Educ Res. 2021;78(4):751–96. [ Google Scholar ] 40. Schunk DH, Zimmerman BJ. Motivation and self-regulated learning: theory, research, and applications Motivation and self-regulated learning: theory, research, and applications. Routledge; 2020. [ Google Scholar ] 41. Ritchie L, Williamon A. Measuring distinct types of performance anxiety in music students. J Res Music Educ. 2022;70(3):296–315. [ Google Scholar ] 42. Pajares F, Schunk DH. Self-beliefs and school success: self-efficacy, self-concept, and school achievement. Perception. 2021;11:239–66. [ Google Scholar ] 43. Locke EA, Latham GP. Building a practically useful theory of goal setting and task motivation: a 35-year odyssey. Am Psychol. 2020;57(9):705–17. [ DOI ] [ PubMed ] [ Google Scholar ] 44. American Psychological Association. Publication manual of the American Psychological Association Publication manual of the American Psychological Association. 7th ed. American Psychological Association; 2020. [ Google Scholar ] 45. Creswell JW, Creswell JD. Research design: qualitative, quantitative, and mixed methods approaches Research design: Qualitative, quantitative, and mixed methods approaches. 5th ed. Sage; 2021. [ Google Scholar ] 46. Fowler FJ. Survey research methods Survey research methods. 5th ed. Sage; 2020. [ Google Scholar ] 47. Dillman DA, Smyth JD, Christian LM. Internet, phone, mail, and mixed-mode surveys: the tailored design method Internet, phone, mail, and mixed-mode surveys: The tailored design method. 4th ed. Wiley; 2021. [ Google Scholar ] 48. Shadish WR, Cook TD, Campbell DT. Experimental and quasi-experimental designs for generalized causal inference Experimental and quasi-experimental designs for generalized causal inference. Houghton Mifflin; 2020. [ Google Scholar ] 49. Devellis RF. Scale development: theory and applications Scale development: Theory and applications. 4th ed. Sage; 2022. [ Google Scholar ] 50. Hamre BK, Pianta RC. Can instructional and emotional support in the first-grade classroom make a difference for children at risk of school failure? Child Dev. 2021;76(5):949–67. [ DOI ] [ PubMed ] [ Google Scholar ] 51. Ryan RM, Deci EL. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. Am Psychol. 2000;55(1):68–78. [ DOI ] [ PubMed ] [ Google Scholar ] 52. Smith BW, Dalen J, Wiggins K, Tooley E, Christopher P, Bernard J. The brief resilience scale: assessing the ability to bounce back. Int J Behav Med. 2021;15(3):194–200. [ DOI ] [ PubMed ] [ Google Scholar ] 53. Schwarzer R, Jerusalem M. Generalized self-efficacy scale. In: Weinman J, Wright S, Johnston M, editors. Measures in health psychology: A user’s portfolio. NFER-Nelson; 2020. p. 35–7. [ Google Scholar ] 54. Tabachnick BG, Fidell LS. Using multivariate statistics. 7th ed. Pearson; 2021. [ Google Scholar ] 55. Field A. Discovering statistics using IBM SPSS Statistics. 5th ed. Sage; 2021. [ Google Scholar ] 56. Hayes AF. Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. 3rd ed. Guilford Press; 2022. [ Google Scholar ] 57. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Routledge; 2020. [ Google Scholar ] 58. Kline RB. Principles and practice of structural equation modeling. 5th ed. Guilford Press; 2023. [ Google Scholar ] 59. Byrne BM. Structural equation modeling with AMOS: basic concepts, applications, and programming. 3rd ed. Routledge; 2021. [ Google Scholar ] 60. Bandura A. Guide for constructing self-efficacy scales. In: Pajares F, Urdan T, editors. Self-efficacy beliefs of adolescents. Information Age Publishing; 2022. p. 307–37. [ Google Scholar ] 61. Niemiec CP, Ryan RM. Autonomy, competence, and relatedness in the classroom: applying self-determination theory to educational practice. Theory Res Educ. 2020;7(2):133–44. [ Google Scholar ] 62. Patrick H, Ryan AM, Kaplan A. Early adolescents’ perceptions of the classroom social environment, motivational beliefs, and engagement. J Educ Psychol. 2021;99(1):83–98. [ Google Scholar ] 63. Jang H, Reeve J, Deci EL. Engaging students in learning activities: it is not autonomy support or structure but autonomy support and structure. J Educ Psychol. 2020;102(3):588–600. [ Google Scholar ] 64. Schunk DH. Self-efficacy and academic motivation. Educ Psychol. 2021;26(3–4):207–31. [ Google Scholar ] 65. Zimmerman BJ, Schunk DH. Self-regulated learning and academic achievement: theoretical perspectives. Routledge; 2020. [ Google Scholar ] 66. Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods. 2021;40(3):879–91. [ DOI ] [ PubMed ] [ Google Scholar ] 67. Shrout PE, Bolger N. Mediation in experimental and nonexperimental studies: new procedures and recommendations. Psychol Methods. 2020;7(4):422–45. [ PubMed ] [ Google Scholar ] 68. Bandura A. Social foundations of thought and action: a social cognitive theory. Englewood Cliffs, NJ: Prentice Hall; 1986. [ Google Scholar ] 69. Pajares F. Self-efficacy beliefs in academic settings. Rev Educ Res. 2020;66(4):543–78. [ Google Scholar ] 70. Hu LT, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct Equ Modeling. 2020;6(1):1–55. [ Google Scholar ] 71. McPherson GE, McCormick J. Self-efficacy and music performance. Psychol Music. 2022;34(3):322–36. [ Google Scholar ] 72. Hallam S. The power of music: A research synthesis of the impact of actively making music on the intellectual, social and personal development of children and young people. International Music Education Research Centre. 2023. 73. Burwell K, Shipton M. Strategic approaches to practice: an examination of the one-to-one pedagogical relationship. Br J Music Educ. 2021;30(1):73–96. [ Google Scholar ] 74. Maxwell SE, Cole DA, Mitchell MA. Bias in cross-sectional analyses of longitudinal mediation: partial and complete mediation under an autoregressive model. Multivar Behav Res. 2021;46(5):816–41. [ DOI ] [ PubMed ] [ Google Scholar ] 75. Hendricks KS. The philosophy of perfectionism: anticipated perfection, personal perfection, and adaptive perfectionism. Psychol Music. 2020;48(2):198–216. [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary Material 1. (57.6KB, docx) Data Availability Statement All data generated and analyzed during the current study are presented in Supplementary File 1. Articles from BMC Psychology are provided here courtesy of BMC ACTIONS View on publisher site PDF (1.3 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top