1. Introduction

The importance of learning engagement in modern education was emphasized, and its direct impact on students' learning outcomes was pointed out. In the field of foreign language education, there is relatively little research on the investment in learning French as a second language for college students. With the application of information technology in foreign language teaching, increasing students' online learning engagement has become the key to improving learning quality. As an important factor affecting learning achievement, the relationship between network self-efficacy and learning engagement still needs further research. This study focuses on undergraduate students majoring in English, aiming to explore the dimensions and structure of French online learning engagement, as well as the relationship between network self-efficacy and learning engagement, providing teaching references for educators.

2. Literature review and research hypotheses

Deeply explored the concept, dimensions, and characteristics of learning engagement in online learning. Learning engagement is defined as the time and energy that students invest in educational activities, including behavioral, emotional, and cognitive dimensions. Online learning investment emphasizes time, experience, interest, learning strategies, critical thinking, as well as learning motivation and behavioral processes. Research has shown that the antecedents of online learning engagement include social patterns, self-regulated learning, digital natives, situational interests, positive attitudes, self-efficacy, task value, and teacher presence, while the outcome variables are positively correlated with academic performance and satisfaction.

3. Research methods

The research subjects were undergraduate English majors from 5 universities, with a total of 516 questionnaires distributed. 471 questionnaires were valid, with an effective rate of 91.3%. The average age was 20.18 years old, and female students accounted for 77.3%. The research tool is the adjusted "Online Learning Engagement Scale", which includes four dimensions of cognitive, emotional, participation/interaction, and performance engagement, using the Likert Seven Level Scale. Data analysis was conducted using SPSS 23.0 and Smart PLS 4.0 for exploratory factor analysis, confirmatory factor analysis, descriptive statistics, and correlation analysis, as well as PROCESS Model 59 to explore the moderating effect of network self-efficacy.

4. Research results

Exploratory factor analysis was conducted on the French Language Online Learning Engagement Scale using principal component analysis and Varimax rotation method to extract three factors: emotional engagement, cognitive engagement, and behavioral engagement. The cumulative explanatory variance contribution rate was 76.17%. Confirmatory factor analysis shows that the reliability and convergent validity of the three dimensions are good, and the discriminant validity between the measurement models is also good. The descriptive statistical results show that the mean values of the three factors, from high to low, are cognitive investment, emotional investment, and behavioral investment, all of which belong to the moderately high level, and there is a significant positive correlation between them. Structural model analysis shows that both emotional and cognitive investment have moderate explanatory power for behavioral investment, while emotional investment has lower explanatory power for cognitive investment. Network self-efficacy plays a moderating role in the relationship between emotional investment and behavioral investment. Its level significantly affects the predictive power of emotional investment on behavioral investment, but does not affect the effect of emotional investment on cognitive investment or cognitive investment on behavioral investment. When both network self-efficacy and emotional engagement are high, learners' behavioral engagement is optimal; On the contrary, when both are low, behavioral engagement is the least ideal.

5. Analysis and Discussion

This chapter provides an in-depth analysis of the intrinsic mechanism of college students' engagement in online learning of French as a second language. Research has found that emotional investment has a significant positive predictive effect on cognitive investment, indicating that positive emotional experiences can promote students to adopt effective learning methods. At the same time, cognitive engagement can significantly positively predict behavioral engagement, indicating that the use of online learning strategies can promote students' behavioral participation. In addition, emotional investment also has a positive predictive effect on behavioral investment, that is, students' interest in French and emotional investment can promote their learning participation and performance behavior. Network self-efficacy plays a significant positive moderating role between emotional and behavioral engagement, indicating that as network self-efficacy increases, the predictive effect of emotional engagement on behavioral engagement also increases synchronously. Teachers should attach importance to students' emotional experiences and online self-efficacy, create a positive learning atmosphere through curriculum design and teaching activities, guide students to build confidence, overcome negative emotions, and stimulate students' cognitive and behavioral engagement by improving their digital skills.

6. Conclusion

This study reveals the internal structural relationship between cognitive, affective, and behavioral engagement in online learning of French as a second language. It is found that affective and cognitive engagement have a positive impact on behavioral engagement, and affective engagement can also predict cognitive engagement. Network self-efficacy mainly regulates the effect of emotional investment on behavioral investment. The study provides empirical support for understanding online learning engagement, but further longitudinal research and interviews are needed to explore the dynamic mechanisms.