Based on the studies conducted by Kiss and Lin (2016) and Northcote (2009), it has been indicated that there exists a positive relationship between TB and teaching behaviors. Furthermore, the research conducted by Bereczki and Karpati (2018) suggests that the successful implementation of creative teaching in the classroom by teachers depends on their personal beliefs. Keys (2007) emphasizes that both teachers’ beliefs and habitual ways of doing things can influence their practices in the classroom. Additionally, Choi and Ramsey (2009) propose that practical teaching experiences can assist novice teachers in cultivating appropriate positive beliefs, attitudes, and teaching abilities. The research by Hall and Hord (2011) also demonstrates that true behavioral change in teachers is only possible when they genuinely embrace certain beliefs deep within themselves. Hence, this study proposes Hypothesis 1: Chinese university music teachers’ TB have a significant positive impact on CTB.

Ajzen and Madden (1986) proposed a theory suggesting that beliefs can influence behavior. Subsequently, teachers’ TB have been recognized as a key factor influencing their instructional practices (Ng et al., 2010). Research by Taimalu and Luik (2019) found that constructivist TB positively affect teachers’ technological integration, while transmission TB have a negative impact on technological integration. Teachers with constructivist TB are more inclined to use technology to support student learning, whereas those with traditional TB tend to use technology to support their own curriculum control (Ertmer et al., 2012). Zou et al. (2022) examined the impact of preservice teachers’ TB on their intention to empower student learning using technology from both constructivist and transmission perspectives, with results indicating a positive effect in this research pathway. Therefore, this study proposes Hypothesis 2: Chinese university music teachers’ TB have a significant positive impact on TA.

Research indicates that teachers’ TA is associated with their adoption of innovative teaching methods and materials (Nikolopoulou and Gialamas, 2016). TA can be understood as the willingness of certain teachers to pioneer the adoption of innovative educational technologies and methods within the school environment (Agyei and Voogt, 2014; Teo, 2014; Daniels, 2015). Teachers with high TA are adept at utilizing information and communication technologies as educational tools, seamlessly integrating content and instructional methods. Consequently, they are more inclined to apply ICT to address instructional challenges and select appropriate teaching methods. The extent to which teachers accept innovative technologies also influences their willingness to use these technologies in their daily teaching practices. Teachers who are more receptive to new methods are more effective in introducing new pedagogical approaches, thereby facilitating the dissemination of ICT innovative teaching (Chou et al., 2019). Chou et al. (2019) found a significant positive correlation between TA and the use of ICT for CTB. Therefore, this study proposes Hypothesis 3: TA among Chinese university music teachers has a significant positive impact on CTB.

Van Twillert et al. (2020) found that teachers’ TB constitute a significant belief that directly predicts their behavioral intentions. Additionally, scholars have observed that different TB have varying degrees of influence on behavioral intentions. Bahçivan et al. (2018) investigated the relationship between TB and attitudes among 1,028 teacher candidates and found significant effects of both types of TB on attitudes, with constructivist TB predicting attitudes more strongly than transmission TB. However, differing viewpoints have been proposed as well. Liu et al. (2017) argued that teachers’ constructivist TB significantly influence attitudes, while transmission TB have no significant impact on attitudes. Moreover, scholars have noted that different attitudes have varying degrees of predictive power on behavioral intentions (Wixom and Todd, 2005). Broadly defined, behavioral attitudes are understood as the willingness of individuals to use information technology, use related materials (such as digital textbooks or electronic whiteboards), or engage in teaching practices (Scherer et al., 2018).

Ajzen and Madden (1986) confirmed that behavioral attitudes are more effective predictors of behavioral intentions than attitudes toward objects. In this study, behavioral attitudes refer to teachers’ attitudes toward the use of technology to empower teaching practices, which are expected to strongly predict behavioral intentions. Zou et al. (2022) conducted a study with teacher candidates from seven universities in Zhejiang Province, distributing questionnaires through the online platform “Questionnaire Star,” with a total of 1,120 valid responses. The study found that the behavioral attitudes of teacher candidates played a fully mediating role in the relationship between TB and behavioral intentions. Therefore, this study proposes Hypothesis 4: TA among Chinese university music teachers serves as a significant mediating factor in the relationship between TB and CTB.

Based on the discussions above, this study aims to investigate the influence relationships among TB, CTB, and TA among Chinese university music teachers. The research framework is designed as depicted in the Figure 1.

In this study, an online questionnaire survey was conducted using the TB scale, CTB scale, and TA scale among Chinese university music teachers. A total of 3 Chinese universities were selected, including one comprehensive university, one applied university, and one vocational and technical university. Comprehensive universities typically offer a variety of disciplines and majors, including arts, sciences, engineering, medicine, law, etc., and possess the authority to grant complete undergraduate, master’s, and doctoral degrees. Applied universities focus on cultivating applied talents, emphasizing the integration of theoretical knowledge and practical application. They usually offer majors closely related to social demands, such as engineering technology, management, design, etc. Vocational and technical universities mainly train talents with technical skills, emphasizing the cultivation of practical skills. The majors offered are often related to specific vocational skills, such as mechanical manufacturing, automotive maintenance, catering management, etc.

Convenience sampling was employed in the survey, which was conducted in two phases. This study collected questionnaires online. Both batches of samples were collected using convenience sampling methods and distributed via the online platform “Questionnaire Star.” During the pilot phase, a total of 150 questionnaires were distributed to analyze the item analysis, exploratory factor analysis, and reliability analysis of the scales used in this study. Ultimately, 136 valid pilot questionnaires were collected, resulting in an effective rate of 90.667%.

During the formal survey, questionnaires were distributed with the assistance of colleagues, friends, and through collective activities such as local teacher training. A total of 480 questionnaires were distributed during the formal survey phase. During the data collection process, questionnaires with completion times less than 10 min or incomplete responses were excluded. Eventually, 458 valid questionnaires were collected. When using the questionnaire survey method, the researchers explained the research content, purpose, and requested the participants to sign informed consent forms. Participants were informed that their responses would be anonymous and they could withdraw from the survey at any time.

This study primarily utilizes statistical analysis software such as SPSS and AMOS to validate the proposed theoretical model. The specific research contents include descriptive statistical analysis, common method bias testing, Pearson correlation analysis, structural equation modeling (SEM), and mediation effect testing.

Regarding gender, there were 233 male participants, accounting for 50.900%, and 225 female participants, accounting for 49.100%. In terms of education level, 102 participants had a bachelor’s degree, accounting for 22.300%, 211 participants had a master’s degree, accounting for 46.100%, and 145 participants had a doctoral degree, accounting for 31.700%.

As self-reporting was utilized for data collection in this study, there might be a possibility of common method bias. Therefore, both procedural control and statistical tests were employed to address this issue. Firstly, strict procedural control measures were implemented during the questionnaire administration process, emphasizing that the results of the survey were solely for academic research purposes and ensuring absolute confidentiality of the data provided, with responses being anonymous. Subsequently, to examine whether variables exhibited common method bias, an exploratory factor analysis was conducted for Harman’s single-factor test. If a single factor explains more than 50% of the variance for all variables, it indicates significant common method bias. The analysis revealed that, prior to factor rotation, 12 factors with eigenvalues greater than 1 were extracted. The first factor accounted for 28.955% of the variance, which was less than 50%. Hence, it was concluded that there was no significant common method bias in this study.

To understand the relationships between predictor variables and to avoid multicollinearity issues, Pearson product–moment correlations were used to explore the relationships among TB, CTB, and TA. The correlations between variables are presented in Table 1. All constructs exhibited significant positive correlations. Moreover, the correlation coefficients between predictor variables were all <0.700, indicating moderate correlations, suggesting the absence of severe multicollinearity issues. Therefore, further examination of the relationships between variables using SEM can be pursued.

In this study, a total of 458 valid samples from the formal questionnaire were included and subjected to SEM for validation. Maximum likelihood estimation was utilized for estimation. The model fit indices were obtained to assess the model’s goodness of fit. A smaller χ² value indicates a higher model fit (Jöreskog, 1970); the χ²/df ratio should not exceed 5 (Bentler and Bonett, 1980). The GFI should exceed the recommended value of 0.900, while the AGFI should surpass 0.800. Additionally, the RMSEA and SRMR should be less than the recommended value of 0.080. The CFI should exceed 0.800, and the NFI should surpass 0.800. The results, as shown in Table 2, indicate that all model fit indices meet the standards.

In this study, SEM was employed to investigate the relationships among TB, CTB, and TA, as well as to examine the mediating role of TA and validate the hypotheses proposed in this study.

As shown in Table 3, the standardized path coefficient from TB to CTB is 0.432, with p < 0.001, indicating a significant positive impact of TB on CTB. This suggests that stronger TB are associated with higher levels of CTB, thus validating Hypothesis 1.

Similarly, the standardized path coefficient from TB to TA is 0.517, with p < 0.001, revealing a significant positive influence of TB on TA. This supports the validation of Hypothesis 2, indicating that stronger TB leads to higher levels of TA.

Furthermore, the standardized path coefficient from TA to CTB is 0.236, with p < 0.010, indicating a significant positive impact of TA on CTB. Therefore, Hypothesis 3 is validated, suggesting that higher levels of TA are associated with increased levels of CTB.

To examine the stability of the mediating effect, this study employed the Bootstrap method based on the recommendations of Preacher and Hayes (2008). Specifically, 5,000 bootstrap samples were drawn from the data-set to estimate the coefficients of the mediating effect with a 95% confidence interval.

As shown in Table 4, the results of the analysis confirmed the direct effect of TB on CTB (β = 0.432, p = 0.001). The indirect effect 1, representing the pathway from TB to TA to CTB, had an effect value of 0.122, with a 95% confidence interval ranging from 0.019 to 0.215. Since this confidence interval does not include zero, it suggests that TA mediates the relationship between TB and CTB, thus further validating Hypothesis 4.

This study reveals the mechanism of the influence of teachers’ teaching beliefs on creative teaching behavior. Creative teaching behavior is crucial for cultivating innovative talents, and teachers’ teaching beliefs are important factors influencing their behavior (Northcote, 2009; Kiss and Lin, 2016). Through the mediating role of technological acceptance, this study delves into the intrinsic mechanism of how teachers’ teaching beliefs transform into creative teaching behavior. This finding is of significant importance for guiding teacher education, promoting teachers’ professional development, and improving the teaching quality of music disciplines.

This study explores the impact of Chinese university music teachers’ teaching beliefs on creative teaching behavior and examines the mediating role of technological acceptance. This research not only has theoretical significance but also profound implications in practical applications. Teachers’ teaching beliefs not only affect their choice of teaching methods and means but also directly relate to students’ learning outcomes and the quality of talent cultivation (Berger et al., 2018). Therefore, by guiding teachers to establish scientific teaching beliefs, it is possible to promote teachers to continuously update teaching methods, improve teaching quality, and cultivate more high-quality talents in the field of music.

When conducting music teaching, teachers should select suitable music technology tools based on teaching objectives. For instance, if the goal is to enhance students’ music composition skills, digital music production software can be used. If the objective is to reinforce students’ music theory knowledge, interactive music theory learning platforms may be preferable. To ensure that teachers can fully utilize music technology tools, schools need to provide necessary technical training, which may include software operation guidance and sharing of music production techniques. As music technology continues to evolve, teachers need to continuously update their knowledge and skills to ensure that teaching content remains current. Furthermore, it is essential to pay attention to the potential applications of new technologies in teaching and promptly integrate them into instruction.

To reinforce teachers’ TB, universities can organize a series of teaching training activities to help teachers clarify their teaching philosophies and methods and enhance their teaching skills. Through training, teachers can gain a deeper understanding of their teaching practices. Regularly organizing activities to commend outstanding teaching and inviting experienced teachers to share their teaching experiences and insights can also be beneficial. Through exchanges and discussions, teachers can engage in deeper reflection and understanding of their teaching practices. Additionally, universities can establish teaching incentive mechanisms to recognize teachers who demonstrate excellence in teaching practice. Through incentives, teachers can be encouraged to remain steadfast in their TB and actively engage in teaching activities.