The application of integrated writing tasks which are more demanding in comparison to independent writing activities has gained traction, and research on these tasks tends to augment. However, few studies attempted to examine the role of self-efficacy in high school students' integrated writing performance in an English as a Foreign Language context. To fill this niche, the current study investigates the predictive role of L2-integrated writing self-efficacy in the summarization task. The participants of this study were 131 high school students who were required to complete an L2 reading-to-write task and completed a Likert scale measuring L2 writing self-efficacy beliefs. The results of structural equation modelling (SEM) revealed that self-regulatory efficacy and discourse synthesis efficacy did not significantly predict summarization performance unlike writing conventions self-efficacy. The findings also reveal that there is a need for an L2-integrated writing self-efficacy scale for high school students in English as a Foreign Language contexts. Implications for teachers and researchers are discussed.