Internet Learning Volume 3, Number 2, Fall 2014 | Page 99

Internet Learning
Figure 20. Comparison of Renlit and Kerrad Threads – spread.
authors, at the same time in each thread
(points B and E for Jakata, and points C
and G for Naya). Despite the similarity of
the interventions, the subsequent values for
topicSpread, knowledgeActivity, and DiscussionRank
are distinct for each thread.
Figure 19 shows an increase in
knowledgeActivity subsequent to Jakata’s
question at B, with no change after the partner
post at E.
Figure 20 shows topicSpread increasing
to Level 4/Expand after Naya’s
question at C, but no change after the partner
post at G.
As a final point of comparison, we
can use discussionRank to assess the generative
influence of individual questions
on subsequent discussion (see Figure 11
for an explanation of how to calculate discussionRank).
For example, Jakata’s discussionRank
score is 7 at point B, and 3 at
point E. The differences in knowledgeActivity,
topicSpread, and discussionRank values
for Jakata’s questions at B and E signal
some variation in influence, even given the
similar instructional questioning strategy.
There could be many reasons that similar
interventions in similar contexts would
produce varying results. In the case of the
Kerrad thread, Kerrad expresses initial apprehensions
about statistics and analytics.
As a result, the responses from the rest of
the group are focused on helping Kerrad
to understand analytics in the context in
which they were presented. By contrast, the
Renlit thread is more focused and technical
in nature. The Kerrad conversation remains
more static at a level of explanation,
whereas Renlit’s thread shows more change.
The ability to perceive such trends and distinctions
in conversations using a set of familiar
metrics could help instructors more
effectively engage with, assess, and support
learners in online social spaces.
98