Notes: Kanuka & Anderson. Online Social Interchange, Discord, and Knowledge Construction

Kanuka, H. & Anderson, T. (1998). Online Social Interchange, Discord, and Knowledge Construction. Journal of Distance Education. 13 (1) pp. 57-74.

This study presents the results of an exploratory multi- method evaluation study and transcript analysis of an online forum. The researchers used a constructivist interaction analysis model developed by Gunawardena, Lowe, and Anderson (1997)¹

Gunawardena et al.’s phases:

1. Sharing/comparing of information
2. Discovery and exploration of dissonance or inconsistency among the ideas, concepts, or statements advanced by different participants.
3. Negotiation of meaning and/or co-construction of knowledge.
4. Testing and modification of proposed synthesis or co-construction.
5. Phrasing of agreement, statenient(s), and applications of the newly constructed meaning.

Method:

The research study focused on the analysis of data obtained from participants in the online forum. We read postings, but did not participate in the forum. At the end of the two-week forum, an online survey was distributed to all participants and a transcript analysis was undertaken. Finally, a telephone survey was conducted with a stratified sample of participants.

Results:

the forum was perceived by the participants as successful in providing opportunities for reflection and exposure to multiple perspectives on topics that were relevant to the participants. There seemed less agreement, however, with the notion that the forum provided opportunity for application of new knowledge and deeper understanding of the issues.

Transcript analysis:

The unitizing process involved a coding operation that separated the participants’ online interactions (postings that fell in phases I through V) from other postings, such as the moderator’s summaries or other general announcements.

The transcript analysis procedure consisted of reading each message and assigning it to one or more phases. A message that contained two or more distinct ideas or comments was coded in two or more phases (the messages were coded independently by both researchers). Discrepancies were discussed, and a single coding was determined from these discussions.

So many of the messages wound up being categorized as Level 1 (basic knowledge) that a grounded theory method was added to the study.

Grounded theory provided a useful collection of strategies (such as constant comparison and analytic meaning) when little is known about a phenomenon—as was the case in this study where the focus was to investigate knowledge construction and social interaction in an online environment. Using grounded theory, we reassessed and then recategorized the postings.

DN: not sure what the addition of grounded theory does to the study. seems like they just threw up their hands, said “WTF?” and fell back on divining chicken entrails to get something out of the transcripts…

These two new categories were generated from the data: social interchange² and social discord and knowledge construction³.

DN: the study turns out to not be directly applicable, but points to some things to watch for when coding - having too many things fall into the basic level or “other” category. How to design the coding to avoid having to fall back on grounded theory?