Procrastination has been increasing since the proliferation of online learning. While traditionally assessed with self-report instruments, online learning offers the possibility to measure objective indicators (log data). In the present study, we aim to find out whether the combination of short scales on procrastination-related traits and log data predict the extent of dilatory behaviour in online tasks and performance (assignment scores). The log data models (which include the number of clicks on the assignment, the interval between thematic block start and first click, and the number of clicks on course activities as predictors) have a better fit and explain more variance than the questionnaire models when predicting delay; and the predictions barely improve when combined. The prediction of performance did not yield any noteworthy effects. Future studies need to diversify predictors by incorporating contextual factors to improve early and/or late predictions and allow classification of dilatory behaviour (e.g. procrastination vs purposeful delay).

自从在线学习的普及以来，拖延一直在增加。在传统上使用自我报告工具进行评估的同时，在线学习提供了测量客观指标（日志数据）的可能性。在本研究中，我们旨在发现与拖延有关的特质的短量表和日志数据的组合是否可以预测在线任务和表现（分配分数）中的扩张行为的程度。日志数据模型（包括作业的点击次数，主题块开始和第一次点击之间的间隔以及课程活动的点击次数作为预测指标）与预测时的问卷调查模型相比，具有更好的拟合度并能解释更多的方差延迟; 合并后的预测几乎没有改善。性能的预测没有产生任何值得注意的效果。