chrisohly_heuristic2

=Heuristic 2=

A constructivist understanding is that learning is a social process. As peer teaching and learning is valued, special provision needs to be made within online courses for the development of community. Dawson (2006), cites several studies, (Bielaczyc & Collins, 1999; Rovai, 2002) which suggest that students’ “sense of community” is a reliable indicator of cognitive learning, as well as indicating increased student satisfaction and higher retention rates (Dawson, 2006, p.496). Numerous studies have explored the quality of student discussion postings by coding keywords and measuring the number of posts and message length.

Dawson’s 2006 study, //Online forum discussion interactions as an indicator of student community,// explores how community can be measured in a large scale quantitative trial. The study suggests that an accurate measurement is formed, not by the quantity of discussion forum postings, but by the number of different student-to-student interactions. The content of the discussion postings is not the focus of this type of monitoring. Instead, three kinds of interactions are identified, student-to-student, student-to-instructor and system postings, which include orphaned posts or those that get no reply. Strong sense community can be measured by the number of student-to-student interactions and if there are many system postings a sense of community declines (Dawson, 2006). Dawson suggests that individual monitoring of students would be inefficient and impossible with a large number of students. The data mining technology used can be aggregated and plotted into a user-friendly format. When data can be visualized, the unit instructor receives a snapshot of a period of time, which shows the number of links between learners in discussion postings. This provides a very clear picture of student involvement with their peers, and aids in identifying students at risk. Instructors can see when and where to intervene and adjust their teaching.

This area, which Dawson terms ‘academic analytics,’ is a relatively new one. Although data is very easy to read once it has been visualized, the process of taking the raw data and turning it into a form that is readily usable by academics is not without challenges. Currently, we have basic tools available within our LMS that can provide very basic reporting within a topic, but no visual generation of data can be made. Dawson, in a 2008 study, categorises the LMS tools into four categories, Administration, Assessment, Content and Engagement. In that study the use of tools within the engagement category (i.e. discussion forum, chat, mail, who’s online) is much higher than the student use in other categories. This is a similar trend to the uni where I work. For us, student usage statistics play an important role in identifying and analysing student behaviour. Regular reporting on tool usage in the LMS provides data on current trends, which is an important indicator of changes over a period of time. This not only enables timely management of IT resources to cater for the changing trends in student usage but also as Dawson (2008) indicates, can provide the faculty and individual teaching unit with data to inform teaching practice.

The social network software, described by Dawson (2006, 2008) was developed by Aneesha Bakharia and Shane Dawson with script developed using Greasemonkey, a Mozilla Firefox browser extension (Dawson, 2008). The software is freely available at this link []