Longitudinal Data Analysis of Web Discussions in a Blended Course
Abstract: Educational data mining/learning analytics community encourages e-teachers to analyze their course data. However, in our opinion, longitudinal course data analysis has not received the attention it deserves. In this follow-up paper we extend our earlier longitudinal studies of a blended course (from the years 2002-2008), but now emphasis on Web discussions. First, we analyze the postings of 40 students, altogether 588 comments, during the last three years 2009-2011. Our results support the findings of the earlier paper underlining the freedom of the students to choose an own topic in the discussions. Secondly, we analyze the types of student postings during 2009-2011 and compare them to the use in 2004-2006. This reveals interesting insight whether the current "more digital native" students have different Web discussion skills compared to the earlier students. Our results hint that remarkably little change has taken place in the abilities of our students in argumentative discussion.
Presider: Richard Ladyshewsky, Curtin Graduate School of Business, Curtin University