Honing in on social learning networks in MOOC forums: examining critical network definition decisions
Alyssa Friend Wise, Yi Cui, Wan Qi Jin
Seventh International Learning Analytics & Knowledge Conference (LAK '17)
Network design
Nodes
Five hundred sixty-seven participants in a MOOC on statistics in medicine.
Ties
A forum relationship defined in five alternative ways, ranging from a direct reply to shared presence in the same discussion thread.
Methods
Automated content classification of 817 threads, five network tie definitions, and comparisons at network, community, and individual-node levels.
Source: ACM Digital Library
Reviewed summary
MOOC forums contain replies about course ideas as well as social, logistical, and technical exchanges. The study asked whether these forms of interaction should be analyzed as one network and how five common definitions of a forum tie affect the resulting structure.
The researchers classified 817 threads and built content-related, non-content, and unpartitioned networks from 3,124 posts by 567 participants. Content and non-content networks differed at network, community, and node levels; most tie definitions were less consequential, except Total Copresence, which produced distinct and potentially misleading status patterns.
SNA
How SNA was used
The study constructed separate interaction networks with Direct Reply, Star, Direct Reply plus Star, Limited Copresence, and Total Copresence ties. It then compared network-level structure, communities, and individual positions across content-related, non-content, and combined forum activity.
Source
Proceedings of the Seventh International Learning Analytics & Knowledge Conference, 383-392
DOI: 10.1145/3027385.3027446
