Connecte Dness

  • Subscribe to our RSS feed.
  • Twitter
  • StumbleUpon
  • Reddit
  • Facebook
  • Digg

Tuesday, 16 August 2005

Social Network Analysis in an Imperfect World

Posted on 13:30 by Unknown
One of the rules of thumb you hear in SNA circles is that you need 80% response rate to have a reasonably valid survey. Having used that rule for a while without asking why, I am now glad to see some explanation. A new paper in press by Borgatti, Carley, and Krackhardt shows how different measurements of network centrality degrade with imperfect data.

For example, suppose you want to know who are the top ten influential people in an organization. Naturally, you decided to measure influence with social network analysis. The problem is that you don't get all the data. What are the chances that any of the actual top ten influencers are included in your calculated list?

With a few wonderfully simplifying assumptions, BCK comes up with a rough and ready answer. If you are missing just 5% of the network data, chances are that your list of ten influencers has three rogues in it. Let 10% of your data slip through your fingers and your list of ten is probably just better than half right. If you're missing 25% of your data, then most of your "top ten" list is really just a selection of the relatively central influencers, and not a top ten at all.

For anyone fond of the veneer of precision SNA puts on fuzzy questions, these results are an equally precise grain of salt. For another dose of context, see here too.
Email ThisBlogThis!Share to XShare to FacebookShare to Pinterest
Posted in | No comments
Newer Post Older Post Home

0 comments:

Post a Comment

Subscribe to: Post Comments (Atom)

Popular Posts

  • Even with Web 2.0, we still occasionally need to meet face-to-face
    [In case my irony did not come through in the subject line, let me preface this post with a comment that I am an online community skeptic. H...
  • How to build your network by Brian Uzzi and Shannon Dunlap
    Last week I analyzed the introductions underlying my professional network. Coincidentally, my colleague Steve Frigand sent me a nice foll...
  • Viewing network data in Excel... with banana
    Today I received an invitation from Harvard's Program on Networked Governance to watch Marc Smith demonstrate the powers of . NetMap -...
  • Why math will rock your world (BusinessWeek)
    Click on the image below to read the latest cover story from BusinessWeek : " Why math will rock your world ." When you are ready ...
  • Holiday Special -- The Corrections
    I am just back from Bethlehem, PA, recovering from family time, and settling in for the final countdown to 2005. It's a longish drive fr...
  • Free online network survey utility for Organizational Network Analysis
    Back in December I gave my readers a Christmas present: this free spreadsheet utility for organizational network analysis. Quite a few peop...
  • I hate physicists; Barry Wellman is God
    I attended a talk recently that reminded me of the not-so-hidden rivalry between sociologists and physicists who study networks. Convenientl...
  • Social isolation in America increasing dramatically
    The front page of today's Boston Globe announces " It's lonely out there. " For substantially more detail on this sobering...
  • Qualitative Data, Quantitative Analysis
    Pacey Foster (soon to be professor in the School of Management at UMASS Boston) points me to this essay by H Russell Bernard , "Qualit...
  • Web science, Webwhompers
    I have just unveiled Webwhompers , which bears the fruit of four years of my teaching Web science at Boston University. The site features a ...

Blog Archive

  • ►  2012 (1)
    • ►  June (1)
  • ►  2010 (3)
    • ►  June (2)
    • ►  May (1)
  • ►  2009 (22)
    • ►  December (1)
    • ►  September (2)
    • ►  August (2)
    • ►  July (1)
    • ►  June (5)
    • ►  May (4)
    • ►  March (2)
    • ►  February (4)
    • ►  January (1)
  • ►  2008 (36)
    • ►  December (3)
    • ►  November (2)
    • ►  October (1)
    • ►  September (6)
    • ►  August (4)
    • ►  July (2)
    • ►  June (8)
    • ►  May (4)
    • ►  April (3)
    • ►  February (1)
    • ►  January (2)
  • ►  2007 (42)
    • ►  December (1)
    • ►  November (1)
    • ►  October (2)
    • ►  September (6)
    • ►  August (6)
    • ►  July (5)
    • ►  June (8)
    • ►  May (4)
    • ►  March (3)
    • ►  February (1)
    • ►  January (5)
  • ►  2006 (63)
    • ►  December (4)
    • ►  October (2)
    • ►  September (2)
    • ►  August (3)
    • ►  July (7)
    • ►  June (10)
    • ►  May (10)
    • ►  April (4)
    • ►  March (8)
    • ►  February (6)
    • ►  January (7)
  • ▼  2005 (136)
    • ►  December (11)
    • ►  November (13)
    • ►  October (11)
    • ►  September (9)
    • ▼  August (10)
      • Pros and Cons of Open Source
      • Join the Flock
      • Goings on in the social capital network blog world
      • Read my blog, stay employed
      • Gossip is good--pass it on
      • Social Network Analysis in an Imperfect World
      • Community blog at the Mass Bay OD Learning Group
      • SNA used to capture Saddam Hussein
      • Why we hate HR
      • Centrality: a social network view of leadership
    • ►  July (10)
    • ►  June (10)
    • ►  May (12)
    • ►  April (13)
    • ►  March (15)
    • ►  February (9)
    • ►  January (13)
  • ►  2004 (99)
    • ►  December (9)
    • ►  November (18)
    • ►  October (13)
    • ►  September (16)
    • ►  August (15)
    • ►  July (20)
    • ►  June (8)
Powered by Blogger.

About Me

Unknown
View my complete profile