Abstract

Background

With the widespread adoption of smartphones, mobile phone addiction has increasingly gained prominence among Chinese college students, exerting a profound and detrimental impact on their learning engagement. This study employs self-determination theory as a framework to examine the mechanisms through which mobile phone addiction affects students鈥� learning engagement. Specifically, it examines the mediating effects of physical activity and academic self-efficacy in this relationship.

Methods

This study was conducted from March to June 2024, using the Mobile Phone Addiction Scale, Learning Engagement Scale, Physical Activity Scale, and Academic Self-Efficacy Scale among college students from eight universities in Shaanxi Province. The survey was conducted using the Chinese online questionnaire platform 鈥淨uestionnaire Star,鈥漚nd 4,562 valid questionnaires were finally obtained. SPSS 29.0 and AMOS 29.0 were used for data analysis and structural equation model testing.

Results

The results revealed significant negative correlations between mobile phone addiction and learning engagement (r = -0.434, p鈥�&濒迟;鈥�0.01), physical activity (r = -0.732, p鈥�&濒迟;鈥�0.01), and academic self-efficacy (r = -0.338, p鈥�&濒迟;鈥�0.01). Conversely, there were significant positive correlations between learning engagement and physical activity (r鈥�=鈥�0.335, p鈥�&濒迟;鈥�0.01), and academic self-efficacy (r鈥�=鈥�0.717, p鈥�&濒迟;鈥�0.01). The study鈥檚 hypothesized model demonstrated a good overall fit, with indices including 蠂虏/df鈥�=鈥�4.213, RMSEA鈥�=鈥�0.040, and GFI, AGFI, NFI, and CFI all exceeding 0.90. Mobile phone addiction was found to directly impact learning engagement (point estimate = -0.150, p鈥�&濒迟;鈥�0.001) and indirectly through three mediated pathways: physical activity (point estimate = -0.068, p鈥�&濒迟;鈥�0.001), academic self-efficacy (point estimate = -0.298, p鈥�&濒迟;鈥�0.001), and their chained effect (point estimate = -0.377, p鈥�&濒迟;鈥�0.001).

Conclusions

This research has expanded the theoretical framework and intervention approaches regarding the inter-relationship between mobile phone addiction and learning engagement. Increasing individuals鈥� physical activity levels and fostering their academic self-efficacy offers a means to alleviate the detrimental impact of mobile phone addiction on the learning engagement of college students.

Introduction

As the state places increasing emphasis on the quality of higher education and promotes positive mental health education, scholars at home and abroad have gradually begun to give extensive attention and conduct in-depth research on learning engagement, which reflects the positive learning state of college students. Learning engagement refers to a student鈥檚 fulfilling and enjoyable mental state during the learning process, along with a strong sense of identification with learning and the ability to maintain concentration [1]. It serves as a crucial indicator for assessing the learning atmosphere and teaching quality of a school [2, 3]. Moreover, learning engagement is a significant predictor of students鈥� academic performance and achievements, serving as a crucial measure for assessing the effectiveness of the entire learning process and predicting academic outcomes [4, 5]. Studies have revealed that college students learning engagement is influenced not only by demographic and sociological characteristics [6] but also by factors such as personal achievement goals, academic efficacy, enthusiasm for learning, academic procrastination, motivational beliefs, and negative emotions [7,8,9]. Previous empirical studies on learning engagement have been confined to merely analyzing the relationships between the variables without systematically examining the impact of college students鈥� learning engagement or exploring the mechanism behind its effects. Therefore, this study establishes an explanatory structural model of college students鈥� learning engagement grounded in self-determination theory. This model involves sorting out the influencing factors of college students鈥� learning engagement and exploring the inter-relations and combined effect mechanisms among these factors. Such an approach aids university management in gaining a deeper understanding of college students鈥� learning engagement, fostering the connotative development of higher education, and is instrumental in guiding undergraduate education reform and ensuring the quality of talent cultivation.

Mobile phone addiction, also known as mobile phone dependence, mobile phone use disorder, or problematic mobile phone use, is a new type of addictive behavior caused by an individual鈥檚 uncontrollable and intense use of a mobile phone [10]. Media dependence theory suggests that the more an individual relies on a medium (e.g., a computer, a mobile phone, etc.) to satisfy his or her own needs and to achieve personal goals, the more the medium plays a role and has an influence on their life [11]. The social displacement hypothesis suggests that uncontrolled mobile phone use takes away time from real social activities, which may weaken social connections and induce negative emotional experiences such as alienation and fear of social relationships [12]. Extensive studies have shown that the stronger the individual鈥檚 dependence on the mobile phone, the more negative impacts it can have. Specifically, mobile phone addiction may cause cognitive failure [13, 14], endogenous attention, reduce sleep quality, and induce negative emotions such as anxiety and depression [12, 15]. Currently, there is a lack of direct research exploring the relationship between mobile phone addiction and academic engagement among college students. However, some studies have indicated that individuals with a high level of mobile phone addiction are more inclined to experience academic burnout [16] and academic procrastination behaviors [17], and academic burnout and procrastination behaviors significantly and negatively predicted learning engagement [18], which means that higher mobile phone dependence may lead to a significant reduction in students鈥� learning engagement. Compared to primary and secondary school students in China, college students enjoy relatively greater freedom in their use of mobile phones and are subject to fewer external influences and restrictions. However, if college students fail to resist the temptation and interference posed by their mobile phones during their daily studies, this may result in more detrimental impacts on their learning engagement. Therefore, this study proposes hypothesis H1: Mobile phone addiction can negatively predict college students鈥� learning engagement.

Through a review of related literature and theories, this study concluded that physical activity and academic self-efficacy may be important mediating variables between mobile phone addiction and college students鈥� learning engagement. Physical activity, defined as an activity with a specific intensity, frequency, and duration that improves physical health, serves as both the content and the means [19]. Mobile phone addiction is an adverse psychological and out-of-control behavioral state in which individuals compulsively, dependently, and impulsively use mobile phones at an excessively high frequency, which causes them to neglect communication and interaction with real people and biases their attention and mind toward mobile phone use. It also increases sedentary behaviors like screen behaviors and affects physical activity [20]. Some studies have shown that mobile phone addiction can trigger various negative habits, affecting college students鈥� daily studies and lives and leading to delays and inactivity in their real behaviors [21]. Currently, there is limited research on the impact of physical activity on learning engagement among college students. In a study analyzing the relationship between college students鈥� lifestyles and learning engagement, Zhang Jia et al. pointed out that, compared to those with healthy lifestyles, students in the sub-healthy and less exercise group had lower levels of learning engagement [22], which reflects that exercise may have a positive predictive effect on learning engagement. In another study on curriculum and instruction, researchers found that exercise can stimulate learners鈥� motivation and encourage more active engagement, enriching the learning environment through the exercise stimulation method [23]. Therefore, this study also proposes hypothesis H2: Physical activity mediates the relationship between mobile phone addiction and college students鈥� learning engagement.

Academic self-efficacy is a specific application of self-efficacy theory in education, which refers to an individual鈥檚 beliefs and confidence in their ability to achieve expected academic performance and complete learning tasks during the learning process [24]. Academic self-efficacy positively affects students鈥� flexibility and initiative in problem-solving, influences motivation and learning behaviors, enhances academic performance, and increases students鈥� academic effort [25, 26]. It serves as one of the proximal influences on learning engagement. There are limited studies on the effects of technology dependence, including mobile phone and internet dependence, on college students鈥� academic self-efficacy. However, it is generally believed that over-dependence on technology may negatively impact various aspects of college students鈥� interpersonal interactions and academic performance. For example, Chiu鈥檚 study noted a significant negative correlation between the degree of mobile phone addiction and academic self-efficacy among college students [27]. Additionally, Liu Yansu et al. pointed out that the higher students鈥� academic self-efficacy (especially regarding their learning ability), the less likely they are to develop mobile phone addiction, resulting in better academic performance [28]. In summary, this study proposes hypothesis H3: academic self-efficacy mediates the relationship between mobile phone addiction and college students鈥� learning engagement.

According to Bandura鈥檚 theory of self-efficacy, which aligns with the effects of physical activity [29, 30], experiences of success and failure and physiological and emotional states can significantly influence self-efficacy. Physical activity is an inherently goal-oriented behavior that requires individuals to overcome difficulties. Adherence to physical activity provides individuals with successful experiences in overcoming these challenges and maintaining physical health and subjective well-being, thereby enhancing self-efficacy. Additionally, the results of a longitudinal study demonstrated a significant positive relationship between physical activity and academic self-efficacy, indicating that physical activity positively predicts an individual鈥檚 academic self-efficacy [31].

In the study of Grey et al., the effects of physical activity on students鈥� self-efficacy, emotional state, and academic performance were examined, and it was found through a 12-week experiment that appropriate physical activity could help enhance students鈥� academic self-efficacy and also make their emotional state more positive, which in turn could promote the improvement of their academic performance [32]. Self-determination theory believes that individuals need to pursue self-actualization. For college students, learning is an important way to realize self-worth and pursue personal growth [33]. Combined with the above, based on examining the separate mediating roles of physical activity and academic self-efficacy, this study will also examine the chain mediating role between mobile phone addiction and learning engagement and propose the hypothesis H4: Physical activity and academic self-efficacy play a chain mediating role in the relationship between mobile phone addiction and college students鈥� learning engagement.

Based on the above research background, this study attempts to construct a comprehensive research model of the potential effects of mobile phone addiction, physical activity, and academic self-efficacy on college students鈥� learning engagement (Fig.听1) and to test these hypotheses.

Fig. 1

Hypothetical model of the relationship between mobile phone addiction, physical activity, academic self-efficacy and learning engagement

Methods

Participants

This study used the random sampling method to generate links () through the Questionnaire Star platform and then conducted surveys through Chinese social media platforms such as WeChat and QQ in eight universities, including Xi鈥檃n University of Electronic Science and Technology, Northwestern University, Shaanxi Normal University, and Xi鈥檃n Polytechnic University, etc. Because senior students are busy near graduation, this study only surveyed freshmen, sophomores, and junior students. The survey started on March 24, 2024, and ended on June 2, 2024, lasting 10 weeks. A total of 4,800 questionnaires were collected, and 4,562 valid data points were retained based on the screening principles of 鈥渞everse question test,鈥� 鈥渞egularity,鈥� and 鈥渢oo short a period to fill out the questionnaire.鈥� Finally, 4,562 valid data points were retained as the sample of this study, including the effective recovery rate of the data was 95.0%. Among them, 3,570 (78.3%) were male, and 992 (21.7%) were female; 1,983 (43.5%) were freshmen, 1,473 (32.3%) were sophomores, and 1,106 (24.2%) were juniors; and the age was (19.59鈥壜扁��1.21) years.

Measures

Mobile phone addiction scale

The Mobile Phone Addiction Tendency Scale for College Students, developed by Xiong Jie et al. (2012), was used to assess the degree of individual mobile phone addiction [34]. The scale consists of 16 questions and encompasses four dimensions: withdrawal symptoms, salient behaviors, social soothing, and mood changes. Responses are measured on a 5-point Likert scale, where 鈥�1鈥� indicates 鈥渧ery non-compliant鈥� and 鈥�5鈥� indicates 鈥渧ery compliant.鈥� The scale has a minimum score of 16 and a maximum score of 80; thus, a higher score reflects a greater tendency toward mobile phone addiction. The Cronbach鈥檚 alpha value for the scale was 0.925.

Learning engagement scale

The Learning Engagement Scale [35], revised by Fang Laitan and Shi Kanwas, consists of three dimensions: concentration, dedication, and vigor, encompassing 17 items in total. A 7-point Likert scale is employed, with 鈥�1鈥� indicating 鈥渘ever鈥� and 鈥�7鈥� indicating 鈥渁lways.鈥� Higher scores on the individual dimensions and the total score signify greater levels of learning engagement. The Cronbach鈥檚 alpha for the original and revised scales achieved 0.951 and 0.967.

Physical activity scale

The revised version of the Physical Activity Rating Scale (PARS-3) [36], developed by D.C. Leung, was selected for this study. 36 Physical activity participation was assessed based on three dimensions: exercise frequency, exercise duration, and exercise intensity. The formula used to calculate physical activity is: Physical Activity鈥�=鈥塃xercise Frequency Score 脳 (Exercise Time Score 鈭掆��1) 脳 Exercise Intensity Score. This scale utilized a 5-point Likert scale, with each question offering five response options scored from 1 to 5. The total score ranged from 0 to 100, with higher scores indicating higher levels of physical activity participation. The low-level physical activity group was 鈮も��19 points, the moderate-level physical activity group was between 20 and 42 points, and the high-level physical activity group was 鈮モ��43 points. The Cronbach鈥檚 alpha value scale was 0.812.

Academic self-efficacy scale

Academic self-efficacy was measured by referring to the scale used by Kong Lanlan et al. [37]. A five-item academic self-efficacy scale was designed, including statements such as 鈥淚f I work hard, I can solve my learning difficulties鈥� and 鈥淚 can easily understand new content in my studies.鈥� Participants rated their agreement with each statement on a 5-point Likert scale, where 鈥�1鈥� indicates 鈥渘ever鈥� and 鈥�5鈥� is 鈥渁lways.鈥� The higher the total score, the greater the sense of academic self-efficacy. The Cronbach鈥檚 alpha for this scale was 0.886.

Reliability analysis of the scale

The measurement model鈥檚 validity was ensured by using a standardized evaluation of validity and reliability. 鈥淩eliability鈥� was gauged using three primary indicators: the Cronbach鈥檚 alpha coefficient, the combined reliability (CR) value, and the average variance extracted (AVE). Specifically, Cronbach鈥檚 alpha coefficient鈥� assesses the internal consistency of the questionnaire, while CR integrates the correlation between the items within the questionnaire, and AVE reflects the questionnaire鈥檚 convergent validity. According to the meticulous testing of these critical indicators, all variables exhibited Cronbach鈥檚 alpha coefficients exceeding 0.8, CR values surpassing 0.7, and AVE values above 0.5, surpassing the recommended thresholds of 0.7, 0.7, and 0.5, respectively. Consequently, the measurement model presented in this paper demonstrates a robust level of reliability and excellent convergent validity, thereby ensuring the reliability and precision of this study鈥檚 findings. The outcomes of the reliability test are presented in Table听1.

Table 1 Reliability and validity testing of the scale

Statistical analysis

Using Amos 29.0 software, a structural equation modeling (SEM) framework was employed. The SEM was visually inspected to ensure its suitability for a sample size ten times the number of measured variables. There were 43 variables in this study, and the effective sample comprised 4,562 individuals, which adequately fulfilled the requirements for data analysis. 鈥淪tatistical significance鈥� was established at a P-value threshold of less than 0.05, and the Bootstrap method was employed to test the mediation effect, with the reporting of a 95% confidence interval.

Results

Common method bias test

This study may be affected by common methodological biases because the data were obtained through anonymous questionnaires filled out by respondents. According to Zhou Hao et al. [38], it is crucial to control the data collection procedures, firstly to ensure that the research instrument is a questionnaire suitable for the Chinese college student population, secondly, to emphasize that the results of the survey data obtained in this study are limited to the use of this study, and lastly, through Harman鈥檚 one-way test, this study was further examined for common method bias by using the SPSS 29.0 software. The results showed that there were seven factors with eigenroots greater than 1. The cumulative variance explained by the first factor was 33.47%, less than the critical value of 40% [39]. It indicates that there is no serious common method bias in this study.

Analysis of group differences in mobile phone addiction, learning engagement, physical activity, and academic self-efficacy

The results of the Mann-Whitney U-test for gender and the Kruskal-Wallis H-test for grade level (see Table听2) showed that the gender differences in mobile phone addiction, learning engagement, physical activity, and academic self-efficacy are significant (p鈥�<鈥�0.05), with girls鈥� mobile phone addiction (41.47鈥壜扁��10.37) higher than boys鈥� (40.02鈥壜扁��12.04), and girls鈥� learning engagement (73.22鈥壜扁��16.92), physical activity (14.97鈥壜扁��11.92), and academic self-efficacy (16.86鈥壜扁��3.12) all significantly (p鈥�<鈥�0.12) different. 12.04), and girls鈥� learning engagement (73.22鈥壜扁��16.92), physical activity (14.97鈥壜扁��11.92), and academic self-efficacy (16.86鈥壜扁��3.12) were lower than boys鈥� (76.27鈥壜扁��17.99, 16.28鈥壜扁��13.14, 17.55鈥壜扁��3.43). Grade-level differences in mobile phone addiction, learning engagement, physical activity, and academic self-efficacy were not significant (p鈥�&驳迟;鈥�0.05).

Table 2 Mann-Whitney U test for gender and Kruskal-Wallis H test for grade level (惭鈥壜扁赌塖顿)

Descriptive statistics and correlation analysis

The descriptive statistics and correlation results of the variables are shown in Table听3. The results showed that mobile phone addiction was significantly negatively correlated with learning engagement, physical activity, and academic self-efficacy (r =-0.434, p鈥�&濒迟;鈥�0.01;r =-0.732, p鈥�&濒迟;鈥�0.01r =-0.338, p鈥�&濒迟;鈥�0.01) and that there was a significant positive correlation between learning engagement, physical activity, and academic self-efficacy (r鈥�=鈥�0.335, p鈥�&濒迟;鈥�0.01r鈥�=鈥�0.717, p鈥�&濒迟;鈥�0.01); physical activity was significantly positively correlated with academic self-efficacy (r鈥�=鈥�0.315, p鈥�&濒迟;鈥�0.01). Gender was significantly positively correlated with mobile phone addiction (r鈥�=鈥�0.051, p鈥�&濒迟;鈥�0.01) and significantly negatively correlated with learning engagement, physical activity, and academic self-efficacy (r = -0.071, p鈥�&濒迟;鈥�0.01; r = -0.042; p鈥�&濒迟;鈥�0.01; r = -0.084, p鈥�&濒迟;鈥�0.01). The results of the correlation analysis were consistent with the theoretical expectations, providing a feasible analytical basis for subsequent hypothesis testing.

Table 3 Mean and correlation between variables

Analysis of the mediating effect of physical activity and academic self-efficacy

To effectively control measurement error and test the mediating role of physical activity and academic self-efficacy between the independent variable of mobile phone addiction and the dependent variable of learning engagement, this study used the method of structural equation modeling to test the chained mediation effect. The essence of structural equation modeling estimation is covariance structure analysis, that is, comparing the difference between the expected covariance matrix of the hypothetical model of the study and the covariance matrix of the actual sample data. If the difference is within the appropriate range, it indicates that the hypothetical model fits well with the actual data and verifies the reasonableness of the hypothetical model. According to the mediation effect test process proposed by Wen Zhonglin et al. [40], the fit indices of this study鈥檚 model after correction are shown in Table听4, which shows that in the overall model fit test, the value of the chi-square degrees of freedom ratio is 4.213, which is less than 5.000, indicating that the fit between the hypothesized model and the sample data meets the requirements; the value of the RMSEA is 0.040, which is less than 0.080, which indicates that there is a good fit; the rest of the relevant indicators are also greater than the standard value of 0.900. It indicates that the fit of the sample data and the hypothetical model of this paper is better, and the model can be used for further empirical analysis.

Table 4 Model fit indices

Fig. 2

Path analysis results of mobile phone addiction, learning engagement, physical activity and academic self-efficacy. Note: The coefficients in the graphs are standardized; ^***p鈥�&濒迟;鈥�0.001

Analysis of the mediating effect of physical activity and academic self-efficacy

To validate the mediating roles of physical activity and academic self-efficacy, we employed the bias-corrected non-parametric percentile Bootstrap method to test their mediation effects between mobile phone addiction and learning engagement. With a Bootstrap sample size of 5,000, 95% confidence intervals (CIs) were computed. If the 95% CI for the standardized path coefficients does not include zero, it signifies a significant mediation effect. The results displayed in Table听5 indicate that the 95% CI for the mediation path from mobile phone addiction to learning engagement through physical activity is [-0.122, 鈭掆��0.048], confirming a significant mediation effect and validating research hypothesis H2. Similarly, the 95% CI for the mediation path through academic self-efficacy is [-0.746, 鈭掆��0.141], indicating a significant mediation effect and validating research hypothesis H3. Furthermore, the 95% CI for the mediation path involving both physical activity and academic self-efficacy is [-0.110, 鈭掆��0.080], demonstrating a significant mediation effect and validating research hypothesis H4. Upon further analysis of the effect sizes of each variable on learning engagement, it was found that the total effect of mobile phone addiction on learning engagement is -0.527, with a direct effect of -0.150 and a total indirect effect of -0.377. The mediation effects for the three paths鈥攎obile phone addiction 鈫� physical activity 鈫� learning engagement, mobile phone addiction 鈫� academic self-efficacy 鈫� learning engagement, and mobile phone addiction 鈫� physical activity 鈫� academic self-efficacy 鈫� learning engagement鈥攁re 鈭掆��0.068, -0.298, and 鈭掆��0.011, respectively. The relative mediation effects (i.e., the proportion of the mediation effect to the total effect) are 12.9%, 56.5%, and 2.1%, respectively. Notably, the 95% CIs for all indirect effects exclude zero, suggesting that all three paths are statistically significant.

Table 5 Chain mediated path effect test for mobile phone addiction and learning engagement

Discussion

Analysis of group differences in college students鈥� mobile phone addiction, learning engagement, physical activity, and academic self-efficacy

The Mann-Whitney U test for gender showed significant gender differences in mobile phone addiction, learning engagement, physical activity, and academic self-efficacy among college students, a result that is consistent with and different from the results of previous studies at home and abroad. Among them, the results of this study showed that female students among college students have significantly higher scores of mobile phone addiction than male students, which is consistent with the results of previous studies [41, 42]. Female students tend to use mobile social media more than male students for making social connections and for positive self-presentation. In terms of learning engagement, this study showed that girls scored lower than boys, which is slightly different from the results of previous studies [43, 44]. The influence of gender factors on students鈥� learning is changing with the development of diversity in teaching and learning styles; at the same time, the specificity of the learning environments in different majors has led to boys demonstrating a stronger interest in learning and a higher level of acceptance. In terms of physical activity, female students scored lower than male students, which is consistent with the results of previous studies [12, 45]. This may be related to the fact that male students鈥� interest in physical activity and sports participation behavior is higher than that of female students. In terms of academic self-efficacy, male students scored higher than female students, which is consistent with the results of previous studies [46]. The reason may be that female students are more susceptible to the influence of other people鈥檚 evaluations of their own learning and have lower self-confidence than male students, which leads to lower academic self-efficacy for female students than male students. Self-determination theory proposes that humans are inherently positive and growth-oriented beings, driven by a natural tendency to pursue psychological development and exhibiting a strong sense of subjective initiative [47]. The positive, proactive, and self-regulating mindset and attitude demonstrated in the individual learning process, which characterizes learning engagement, aligns perfectly with the principles outlined by self-determination theory. Consequently, educators can effectively focus on nurturing college students鈥� interest in learning during the teaching process, thereby encouraging them to engage more actively in their studies, explore with enthusiasm, and seek knowledge with genuine curiosity.

The relationship between mobile phone addiction and college students鈥� learning engagement

The findings of this study indicate that mobile phone addiction significantly and adversely predicts college students鈥� learning engagement, aligning with the outcomes of prior research [17, 48]. This result further supports the media dependence theory [49]. When college students develop excessive dependence on their mobile phones, it results in constant interruptions in their learning processes due to mobile phone information, subsequently disrupting their study plans and tasks.

The Attention Resource Theory, also known as the energy allocation model of attention, provides another explanation for this relationship. It suggests that dividing an individual鈥檚 mental resources among more than two tasks creates a competition for attention resources [50]. Additionally, days with higher levels of mobile phone addiction are more likely to be associated with academic procrastination [17] and burnout behavior [16]. Based on the above illustration, mobile phone addiction has a negative impact on an individual鈥檚 cognition, emotion, and daily study behaviors, and this may be an intrinsic cause of lower learning engagement in college students. Based on these observations, mobile phone addiction exerts a detrimental impact on an individual鈥檚 cognition, emotion, and daily study behaviors, potentially serving as an intrinsic factor behind reduced learning engagement among college students. Given this context, colleges and educators must give heed to the burgeoning issue of mobile phone addiction among students. To effectively mitigate and intervene in such addictive behaviors, it is crucial to deploy strategic interventions encompassing mental health education, imparting training in time management techniques, and establishing a mobile phone-free classroom ambiance. These initiatives are pivotal in fostering heightened learning engagement and augmenting academic achievement among college students.

The mediating role of physical activity between mobile phone addiction and learning engagement

This study reveals that physical activity partially mediates the impact of mobile phone addiction on college students鈥� learning engagement, validating hypothesis H2. This finding aligns with previous research indicating that mobile phone addiction decreases participation in physical activity [21], and physical activity promotes learning engagement [48]. By exploring these three variables, this study confirms that problematic mobile phone use is a crucial factor in the impact of college students鈥� healthy living and learning. College students who are mobile phone addicted or have difficulty quitting mobile phone use tend to shift their attention preference to the virtual online world, resulting in a large amount of remaining leisure time being crowded by screen behaviors and do-as-you-go behaviors, while the reality is more often presenting a passive and lazy state of physical activity engagement. Some studies have shown that individuals with mobile phone addiction are more inclined to find satisfaction in mobile phone use (mobile phone games, online dating, etc.) and tend to respond to real-life events with negative self-presentation and fluctuating emotional performance [51]. Physical activity stands as a vital means of fostering real-life interpersonal interactions and maintaining physical fitness. However, mobile phone addiction has emerged as a significant barrier preventing college students from engaging in physical activity. The intensity of this addiction directly correlates with the potential negativity towards physical activity; the more severe the addiction tendency, the more detrimental its impact on physical activity is likely to be. Notably, physical activity serves as a protective factor in the academic development of college students, and engaging in appropriate levels of physical activity enhances students鈥� dedication to their studies [52]. Therefore, mobile phone-addicted college students can reduce their learning engagement by decreasing their participation in physical activity.

The mediating role of academic self-efficacy between mobile phone addiction and learning engagement

This present study shows that academic self-efficacy mediates the relationship between mobile phone addiction and learning engagement, which validates this study鈥檚 hypothesis H3 and is consistent with the results found in previous studies [53]. Mobile phone addiction causes students to spend a significant amount of time on non-academic activities, which reduces the amount of time and efficiency spent on learning and ultimately affects academic performance. This decline in academic performance further diminishes and weakens the academic self-efficacy of the students. Students with high academic self-efficacy are more likely to be interested in learning tasks and willing to explore and learn on their initiative. Schunk鈥檚 study showed that students with high self-efficacy showed higher levels of learning engagement, motivation, effort, and persistence [54]. However, students with a tendency to become addicted to mobile phones have less self-control, and they do not have clear goals for their education and a reasonable development plan, leading to decreased academic self-efficacy and academic burnout, thus weakening their intrinsic motivation to learn.

The chain mediation of physical activity and academic self-efficacy between mobile phone addiction and learning engagement

This present study also found that physical activity and academic self-efficacy had a chain mediating effect between mobile phone addiction and learning engagement, which verified this study鈥檚 hypothesis H4 and previous studies鈥� findings [55]. Physical activity positively affected students鈥� academic self-efficacy. Physical activity significantly and positively predicted academic self-efficacy, a result that fits Bandura鈥檚 self-efficacy theory, and academic self-efficacy, as a branch of self-efficacy, is also a mediating variable between external stimuli and behavior. Through physical activity, students can make progress in physical fitness and skills, which can enhance their self-esteem and self-confidence, and positively increase their academic self-efficacy. Other studies have also found that regular physical activity significantly improves students鈥� academic self-efficacy and performance [56]. Students who participate in physical activity perform better in terms of academic self-efficacy and academic achievement [57]. In summary, reducing mobile phone addiction among college students can promote participation in physical activity, enhance academic self-efficacy, and indirectly increase their learning engagement.

This study explores the impact of mobile phone addiction on the learning engagement of Chinese college students, revealing the mediating roles of physical activity and academic self-efficacy between the two. It provides practical intervention strategies for college educators, aiding in the design of more effective health promotion programs and academic support projects for Chinese universities. By alleviating mobile phone addiction and enhancing students鈥� learning engagement and performance, this study contributes to cultivating healthier, more proactive, and efficient talents for the long-term development of society.

Conclusion

[1] Physical activity and academic self-efficacy serve as partial mediators in the relationship between mobile phone addiction and learning engagement.

[2] Physical activity and academic self-efficacy function as chain mediators in the link between mobile phone addiction and learning engagement.

Limitations of the study

This present study has some limitations. First, it only explored the mediating mechanism between mobile phone addiction and learning engagement and did not consider individual differences. Future research can further explore the relationship between mobile phone addiction and learning engagement from the perspective of moderating mechanisms. Second, due to the limitations of the cross-sectional research design, this study could only observe the relationship between the variables at a single point in time and could not track or determine the causal relationship. Therefore, future research should consider adopting a longitudinal or experimental design in order to more accurately determine the causal relationships between the variables. Finally, this study confirmed that physical activity and academic self-efficacy play a crucial role in mediating the relationship between mobile phone addiction and learning engagement. However, there may be other mediating factors between mobile phone addiction and learning engagement, such as social support and mental health. Future research will consider incorporating other theories to explore more pathways of influence and relationships.