A collaborative reading annotation system with formative assessment and feedback mechanisms to promote digital reading performance

The collaborative reading annotation system (CRAS) has been proved its success in promoting reading performance in comparison with traditional paper-based reading. However, there is still lack of an effective formative assessment and feedback mechanisms in the CRAS, which can assist learners to promote their self-regulated learning and reflection. Therefore, this study uses C4.5 decision tree to develop a CRAS with formative assessment and feedback mechanisms (CRAS-FAFM) based on four considered social network indicators, which could forecast the learners with low reading comprehension and suggest them to interact with the learners who are predicted with high reading comprehension performance and infrequently interact in the digital reading activity in order to enhance their reading comprehension through interactive discussion. Accordingly, this study intends to discuss the effects of learners who use the CRAS-FAFM and CRAS without formative assessment and feedback mechanisms (CRAS-NFAFM) on reading comprehension performance and interactive discussion. A total of 55 students from two classes of a Taiwan's primary school participated in the experiment. One class of students was randomly assigned as the experimental group using the CRAS-FAFM to support digital reading while the remaining class was randomly assigned as the control group using the CRAS-NFAFM. Analytical results show that the average prediction accuracy rate of the developed CRAS-FAFM in identifying the learners with low reading comprehension performance is as high as 68.33%. Moreover, compared to CRAS-NFAFM, the CRAS-FAFM provides remarkable benefits in promoting the reading comprehension performance and interactive discussion on the discussion level of comparison, discussion, and analysis, particularly for the learners with low prior knowledge.