Summer 2009 – Stanford Media X Workshop: New Metrics for New Media: Analytics for Social Media and Virtual Worlds

Stanford University - Media X Program

I will lead a workshop with Martha Russell on social network analysis of social media as part of the Stanford Media X Summer Institute on New Metrics for New Media: Analytics for Social Media and Virtual Worlds this Summer.  I am looking forward to working with the folks at Media X which hosts a range of cutting edge events devoted to exploring the newest trends in technology and society.


It is also worth noting that the traveling exhibit “Places and Spaces” will be displayed through the MEDIA X program at Stanford until December 18th, 2009.  There is a May 18, 5-6:30pm Reception in Wallenberg Hall on the campus.  The show includes an image I worked on with Danyel Fisher and Tony Capone that represents an overview of Usenet newsgroups.

2005 Usenet Treemap

The show includes a variety of information visualizations and maps that illustrate the utility of graphical representations of complex concepts and terrains.  From the Media X site:

“The Places & Spaces exhibit, at Wallenberg Hall has two components. The physical component is available for display and allows for close visual inspection through high-quality prints. It is meant to inspire cross-disciplinary discussion on how best to track and communicate human activity and scientific progress on a global scale. It includes hand-on science maps for children. The online counterpart provides links to a selected series of maps and their makers along with detailed explanations of why these maps work.” [Link]

There will be a reception following the May 18 Seminar that will include Jeff Heer and students, Katy Borner (virtual presence) and other mapmakers of the Places & Spaces exhibit.

Conference: Web 2.0 in San Francisco – “Beyond Buzz: On Measuring a Conversation” with Kate Niederhoffer

2009 Web 2.0 San Francisco Logo

I will be speaking with Kate Niederhoffer from Dachis at Web 2.0 Expo Wednesday, April 1st at 10.50 in San Francisco.  We will be speaking about:

Beyond Buzz: On Measuring a Conversation

What is the most meaningful way to understand and measure a dialogue? As marketing transforms from a broadcast model to a conversational one, which constructs should be captured and how do you measure them? Is it necessary to make a distinction between the metrics used to tap into the value of a conversation per se and the ROI of a social media marketing campaign?

The proposed presentation offers new strategies to think about and tap into the depth of interactions and emotional connections people have online. Beyond buzz levels, sentiment, and other core metrics typically provided by brand monitoring solutions, the presentation will offer methods to understand a conversation: how emotional is it, how in synch are the constituents, how intimately do they relate to the brand or product? How much trust do the constituents reveal?

Marketing efforts that take advantage of technology to enable community and collaboration render traditional metrics limiting, at best. Traditional metrics have been optimized for more passive exposure to a specific message, frequency of exposure is considered a proxy for relevance; and, the premium is on reach over quality.

Primarily due to its more participative dynamic, a conversation engages constituents unlike static messaging. As many in the industry have noted, a natural development is to measure engagement. However, there is little consensus on what engagement means and how it can be measured. Often it is calculated by merely adding traditional metrics, assuming more is better.

The presentation will introduce new constructs and present case studies with empirical research demonstrating more valuable, still measurable constructs than the core metrics currently in use.

NodeXL – a sample of user goals and research interests as a Wordle cloud

NodeXL - a sample of user goals and research interests as a Wodle cloud
NodeXL - a sample of user goals and research interests as a Wordle cloud

Thanks to inspiration from Adam Perer, here is a picture of some of feedback text — a rough presentation of the ideas, topics and research goals submitted by a collection of attendees of a recent lecture on NodeXL at Harvard.  A recording of the slides and audio from the talk is available from the Harvard Program on Networked Governance website and also on the Government Innovators Network website.

A copy of the video that does not require registration is available here.

As people registered for the lecture some took the time to share their interests in network analysis and the NodeXL project.  The key themes and topics from these statements focus on studying the social network of various populations of interest including (in no particular order): governments, enterprises, migrants, terrorists, innovators, politicians, corporations, criminals, research literature, sick people and online communities.

Research goals were focused on tasks like:

  • Map a complex set of relationships
  • Find key people: bridge roles, sinks, sources, etc..
  • Find “missing people”
  • Measure cohesion of subgroups
  • Map diffusion and change overtime
  • Find network correlates of the adoption of innovation

This feed back will shape the kinds of features we explore in the NodeXL project.  What are your goals for exploring network data sets?  What questions are you trying to answer?  What kinds of network structures are of greatest interest?  How can NodeXL best support that exploration?

Your comments welcome!