Engagement is key to students’ success in online learning, because it’s positively related to academic achievement and is negatively associated with dropping out (Fredricks et al., 2004; Steele & Fullagar, 2009). The dropout rate in online courses is often higher than in traditional classroom settings (Levy, 2007). This highlights the importance of identifying effective strategies to maintain student motivation and engagement in virtual learning environments.

Research suggests that behavioral engagement is a particularly strong predictor of academic achievement and student retention (Fredricks et al., 2004). One of the key indicators of behavioral engagement is students’ homework engagement. Research has shown that homework engagement may contribute to higher academic achievement (Ramdass & Zimmerman, 2011).

Engagement and its components are characterized by malleability and are responsive to variations in the context of learning (Fredricks et al., 2004). Hence it’s important to investigate what groups of contextual factors can help to enhance and foster student engagement. Social-relatedness factors are crucial for learners’ engagement (Lam et al., 2012). In particular, peer interaction has been identified as a key factor in promoting engagement.

Despite the significant attention given by researchers to peer interaction in education, the relationship between peer interaction and homework engagement remains under-researched. Additionally, many previous studies have relied on survey data to measure both engagement and peer interaction, which may introduce response bias and may not accurately reflect students’ actual behavior.

This study aims to address these gaps in the literature by investigating the relationship between homework engagement and peer interaction via social media measured on digital trace data. We collected the digital trace data from a literature course on the online-learning platform. We analyzed the data on students’ homework engagement, peer interaction on social media, and academic achievement. The data were collected over an 11-week period and included 10,846 observations from a total of 318 unique students.

Our results revealed a strong and positive correlation between homework engagement and students' academic achievement. This finding highlights the importance of further investigating factors that may contribute to increased homework engagement, as this variable appears to play a significant role in mediating the relationship between contextual factors and academic success.

Our results provide empirical evidence to support the contribution of peer interaction to student engagement. We also revealed that while peer interaction positively contributes to homework engagement, its absence appears to have a more pronounced impact on homework disengagement. Specifically, when a student is disengaged from peer interaction, the negative impact on their homework engagement seems to be stronger. We demonstrated that peer interaction is important for homework engagement among both low- and high-performing students. However, there are differences in the effect size for these two groups, with the factor of interaction or its absence appearing to be more contributional for high-performing students.

Our findings emphasize the importance of promoting peer interaction, as well as monitoring instances of disengagement from peer interaction and identifying students who consistently refrain from participating in these activities.

Keywords: Academic achievement, engagement, homework engagement, online learning, peer interaction.