Flipped learning (FL) makes greater use of students' self-regulated learning (SRL) skills when they regulate their online learning behavior. Previous research has shown the value of SRL support during FL to enhance students' SRL and learning outcomes. However, as previous studies have indicated that SRL behavior varies, should SRL support be tailored to these differences in SRL? We applied latent profile analysis to identify subgroups in 150 eighth-graders during FL. We used practically relevant online behavioral data to represent students’ online SRL activities, which we gathered unobtrusively in an ecologically valid secondary educational classroom setting. We found five distinct SRL profiles from low completion and no activity to full completion and very high activity. In addition, students in the profile who showed low SRL activity achieved significantly worse learning outcomes than students in the three profiles with higher SRL activity. Finally, we explored whether SRL activity profile membership can be explained by student characteristics (i.e., self-reported SRL, motivation, and prior knowledge). None of the student-level variables predicted profile membership, but our approach offers leads for future research to further investigate the potential of tailored SRL support.

翻转学习（FL）在调节学生的在线学习行为时会更多地利用他们的自我调节学习（SRL）技能。先前的研究表明，FL期间SRL支持的价值对于增强学生的SRL和学习成果很有帮助。但是，正如先前的研究表明，SRL行为各不相同，是否应针对SRL中的这些差异量身定制SRL支持？我们应用了潜在的特征分析，以在FL期间识别150个八年级学生中的亚组。我们使用了实用的在线行为数据来表示学生的在线SRL活动，我们毫不客气地将其收集在具有生态学意义的中等教育课堂环境中。我们发现了五个不同的SRL配置文件，从低完成度和无活动状态到完全完成和非常高的活动状态。此外，与那些具有较高SRL活动的三个配置文件中的学生相比，具有较低SRL活动的配置文件中的学生获得的学习成果明显较差。最后，我们探索了SRL活动档案成员资格是否可以通过学生特征（即，自我报告的SRL，动机和先验知识）来解释。没有一个学生水平的变量可以预测个人档案的成员身份，但是我们的方法为将来的研究提供线索，以进一步研究定制的SRL支持的潜力。