WEEK 4: QUALITATIVE DATA COLLECTION

WEEK 4 CONT’D.

3.   ONLINE DISCOURSE ANALYSIS. The text-based nature of online seminars enables storage or archiving of the text, verbatim, whereby the discussion can be analyzed. This is the basis of online discourse analysis. The transcript provides a verbatim copy of the discourse that can then be subject to analysis of Collaborative Learning and Knowledge Building that occurred, over time, by various participants, roles, and processes. Quantitative data and qualitative data are available from the online transcripts, and can provide unique perspectives on the nature of the discourse. Quantitative data are often most easily obtained and analyzed as system-generated usage statistics, which are available on most forum software.

Qualitative data are easily available as the transcripts of the discourse.   Although few analytical software tools exist as yet to study online discourse, we do have Analytical Frameworks and Processes (Harasim, 2011) whereby we can analyze the communication and collaboration patterns and whether, how, and why these change over time.

Discourse Analysis of Collaborative Learning and Knowledge Building

Both quantity and quality of messages in an online course or community offer important indicators of knowledge building and each should be studied and be used to deepen understanding of the nature of engagement and degree of success. Success here is understood as the continuity of activity, advance or progress of the activity, completion, and user satisfaction. The quantity of messaging should not be taken as a sole indication of success, but nor should it be ignored. Levels of participation (such as number of messages per day, per person, per topic, size of a message and other quantitative measures) are an obvious and important indicator of the pulse of an online community. It is important in assessing the distribution of communication and level of activity, engagement, democratic participation and verbalization in a group.

DISCOURSE ANALYSIS METHODOLOGY:

  1. Select the transcript of discourse to be analyzed
  2. Select the unit of analysis: typically, for initial analysis the unit is a MESSAGE;
  3. Code each message (and encourage/require secondary coders)
  4. Each message in a discussion forum can be analyzed and coded according to key indicators, whether the message is primarily
    1. SOCIAL
    2. INTELLECTUAL/COGNITIVE by discussant

i.    IG

ii.    IO

iii.    IC

iv.    (optional finer codings can be IG-IO, IO-IC, other, etc.)

  1. MODERATING

i.    MIG

ii.    MIO

iii.    MIC

  1. PROCEDURAL
  2. Other (Instructor comment, if a student-led seminar), etc.?????
  3. Create a work space in Excel
    1. Spreadsheet that numbers each message
    2. Spreadsheet columns categorize message by SIMPO (see 4 above)
    3. Spreadsheet can also include categories of message gender, time sent , role, content
    4. Chart absolutes: i.e., total # of messages, total # of IG messages, total # of IO messages, total # of IC messages, total # of messages by discussants and by moderators, total # of messages by day, etc.
    5. Very Important: Analyze and Chart CHANGE OVER TIME, for example:
      1. Self-intros
      2. IG—àIO
      3. IO—àIC
      4. Level of participation

Level of participation by gender or by some other factor

a)         Qualitative Content Analysis:

In this section, describe your setting and way that it was structured or designed, and the purpose (i.e., online seminar, online community).   What were the questions that posed by the moderator? Were they answered? How well? Were some questions unanswered? Were new questions posed? answered? new issues or perspectives identified?

What can be learned from this process? What should be done the same again?  What would a moderator not do or do differently.

b)         Quantitative Content Analysis:

In online environments, content analysis is especially appropriate and insightful into the nature of the communication patterns and interactions.  Since there is an automatically-generated transcript of the interactions, these data can be studied and analyzed to gain insight into the usage or communication patterns in that seminar.

USAGE PATTERNS

Examine the total volume of messages per online week; then analyze,  looking for patterns [i.e., daily volume, or according to time of day—morning, afternoon, night; individual usage patterns; patterns of different kinds of participants (i.e., instructor, TA, moderators, discussants, males vs. females); response patterns in relation to the 3 questions, or other topic; types of discourse processes (questions vs answers; agreement vs disagreement; social vs topical, responses vs new material or perspectives, etc.].

USER INTERACTIONS

Examine the message interactions among participants.  For example, what were the interactions among moderators? Among discussants? Between moderators and discussants? types of discourse processes (questions vs answers; agreement vs disagreement; social vs topical, responses vs new ideas, material or perspectives, etc.]. How equitable was the distribution of communication?  Did some dominate the discussion? Or did most of the discussants participate more or less to the same level or degree?

Present your analysis of the usage patterns and interactions both in graphical form and with attendant discussion and explanation.

Example of the use of an Excel Worksheet



Figure 1. Percentage of Idea Generating, Idea Organizing and Intellectual Convergence.

 

 

Figure 1 shows the rise of the Idea generating (IG) and the Idea organizing (IO) phases, culminating on the seventh day of the whole seminar. Intellectual convergence (IC) emerged on the seventh day when discussants came to a conclusion on the topic.

Advertisements