in the blogosphere

Social Network Analysis as a Tool for Locating Knowledge 

Return to Research Home

Data Collection Overview

Visualization of the SNA:

First Pass

623 Nodes

Second Pass

Includes Blog Rolls of Central Actors

885 nodes

First K-Core

2+ levels of trust, 112 nodes

Second K-Core

3+ levels of trust, 47 nodes

Third K-Core

4+ levels of trust, 33 Nodes

Participant Pool

24 Nodes

Note to participants:

My gratitude is endless.

Many now claim that we is smarter than me and that connected knowledge flows across people and machines as we find the information we need when we need it. But where do we go on the Blogosphere? Out there, people are sharing new knowledge daily - posting, commenting, and vetting what they believe to be true. Information is more current and more publicly defended than what we find in journals or books, but it's in the form of a personal post. Who do you believe? How do you judge?

This was the question I asked in my dissertation research and the findings  surpassed expectations. For my research on shared knowledge architectures, the place to start was at the two most trusted Blog search engines on the net: Technorati and Google Blog Search. Using proprietary search logic for authority (Technorati) and relevance (Google), these engines rank posts. So, reading posts by Bloggers that are most read by others (popularity) is one criteria. In tandem, I turned to the Blogosphere's version of peer review, the Blog Roll of the retrieved links, to determine who they enjoy reading. Many Bloggers list the other Bloggers who they choose to follow and like bread crumbs, I followed the trail.

Thus, this study tapped into the Blogosphere's version of both popularity and peer review to determine trusted "experts" who are writing about:

This convergence of knowledge, identified via search links and related Blog rolls, created 885 data points. The data was based on Google and Technorati searches 50 returns deep, Blog Rolls from all relevant and available links, and then, in a second pass thru the retrieved data, Blog rolls of all central actors returned from the first layer of inquiry. 

Using NetDraw, a social network analysis tool, the research was then able to drill down into trust using an SNA K-Core analysis, an iterative process in which the nodes are removed from the graphs in order of least connected. This tool reveals central actors of the SNA , and finally identified a population of 24 nodes with 3 or more referrals (links and Blog rolls to their site) from other, centrally-trusted nodes. 

Trust, defined by combining popularity and peer referral, helped identify those we might want to follow as the convergence of e-learning, technology and the changing workplace redefines who is an expert and how we can harness their shared knowledge.

IF we is smarter than me, and

IF the whole is truly greater than the sum of the parts,

THEN it is in the connections that knowledge is enriched. Thus, it was the intention of this research study to gather knowledge on effective  shared knowledge architecture via collective inquiry by community-identified experts. How we can best design and support emergent learning in the creation of organizational and shared knowledge?

I asked the experts.

Full dissertation available via UMI (bound) or here at the research home site (large PDF).

Write me? cmcarmean at gmail dot com