Analysis of my Own Facebook Nodes: Dancing + Denim = Mind Blown.

I had full intention to blog about the wonderful Clay Shirky, but what hasn’t been already said about him and his insights about collaborative action. For about 10 minutes, I found myself staring at a nice floral arrangement at the Second Cup I was in, and checking Facebook.

And then I had a flash of Network Nerd motivation, and thought: I wonder how some of the small world concepts ACTUALLY apply to what I could gather from my own Facebook data.

Now, this isn’t meant to be scientific by any means. Facebook won’t reflect the breadth of social circles in our network, and the people have different reasons for “friending”. But I figure it’d interesting to see how it plays out and I’m hoping it will help me understand some of the concepts more tangibly. Here goes nothing…

I identified 10 strong ties in my social circle (F1 to F10) to establish a beginning sample. For these 10 strong ties, if available, I identified how many Facebook nodes they had and the number of mutual nodes we shared. The results were as follows: (total nodes/mutual nodes)

  • F1(248/18) [11 years – Nursing School]
  • F2 (350/86) [11 years – Nursing School]
  • F3 (na/20) [4 years – Policy job]
  • F4 (242/43) [18 years – high school]
  • F5 (258/46) [18 years – high school]
  • F6 (226/52) [15 years – Early U]
  • F7 (676/15) [6 years – early career]
  • F8 (370/46) [15 years – Early U]
  • F9 (244/30) [15 years – early U]
  • F10 (660/61) [18 years – high school]

As you can see, I identified how I know each node to provide the  homophilly and propinquity context. In terms of Small World, homophily is major factor accounting for social circle barriers and structures. On average, my friends (I have time calling them nodes) have 365 nodes, sharing on average, 40 mutual nodes with me. The friends that I have  with the highest mutuality are nodes I have known the longest.

Thus, over time, some of my circles have crossed (e.g. highschool friend meeting university friends). The lowest mutuality were my most recent strong ties, whom I haven’t really introduced to my original circle.

Interestingly, this cursory data supports the idea of safety networks because of the the shared mutuality between actors. The number of mutual nodes does not indicate it’s a safety network; I’m confirming it to be a safety network based on my lived experience with these people. However, what is consistent with network theory is that my safety network predominantly shows a higher number of mutual nodes (compared to what you’ll see below). According to Kadushin, the limitation of safety networks are that they are blackboxed and offer little opportunity for effectance. This higher number of mutuality of nodes indicates a lack of structural holes.

Then I decided to look at my Facebook nodes with the highest node counts (I consider these people my weak ties):

(nodes/mutual nodes)

SN1: 3000/23

SN2: 1300/4

SN3: 1900/3

Like Kadushin identifies, a very small number of people have a very large number of connections. In terms of network theory, these are brokers within my social circle. What caught me off guard is how low the number of mutual nodes was.

But you want to know what’s even more surprising?

Of the the 10 strong ties I’ve identified, only 3 of them were directly connected to these weak ties, and it was only to one of them (SN1). This supports that these weak ties have little to do with my safety network. However, it also illustrates how these weak ties act as “friend-of-friend” links for me and my social group to other social circles and the grand network, as well as bridges for transivity and effectancy.

Now, compared to some of the applicable principles of small world:

1. The average person in the world knows 300 people – Facebook says I know 400, and my strong ties average 365. The more people you know and others know, the great the chance of small world.

2. My weak ties, SN1, SN2, and SN3, demonstrate a power distribution of connections like Kadushin indicates is typical in a network.

3. The number of paths in small world is larger than would be expected – even the permutations between me acting as a friend-of-a-friend (400 nodes) to SN1 (3000 nodes) creates an unfathomable number of paths for my strong ties (which also have 300 nodes).

4. Barriers and structures are social circles on the basis of homophily of interests and attributes. In my social group,  homophilly is related to jobs/school/sports/etc. The homophily I have with SN1, SN2, and SN3? Denim & Dancing – I sold denim with SN1 and took classes with SN2 and SN3

5. Overlaps between structures serve to produce strong world. As a node, I overlap my profesional social circle, my old highschool social circle, my university social circle. Along with my connection to the power-distributed nodes (SN1, SN2, and SN3) and their potential for overlap, it makes small world more likely for me.

Conclusion – while not perfect and definitely up for criticism, being able to apply and see how a sample of my network relates has helped me see how Small World operates and ties in some of the previous concepts from network theory.

UPDATE: I was further trying to tie in concepts from previous chapters to what Facebook has made tangible. This exercise has really enforce our observations about the difficulty of partitioning networks for analysis. Fundamentally, I understand this exercise does not address a lot of the concepts of network theory, but using my Facebook network and analyzing to some degree has made interesting and palatable. With that said, because this was a partitioned look, reconciling and moving between scale feels daunting.

I also realized… we’re connected by dancing and denim somehow.

Do you know who dances in denim? 

Kevin Bacon does. Mind blown.


7 thoughts on “Analysis of my Own Facebook Nodes: Dancing + Denim = Mind Blown.

  1. Dear lord, can you please tutor me on how to apply these concepts that I’m reading? You’ve got it so down pat!

  2. Holy crap Ian! That is quite the analysis! *Mind Blown*
    What I miss about Facebook is the feature where you could note how you know someone e.g. ‘we went to school together, we worked together etc’. I don’t know why Facebook removed it and its too bad they did as it would give insight to network analysis.

    • Thanks Leah! I remember that – Facebook was really trying to emphasize relationships. They used also recommend friends quite often. It doesn’t do that as much anymore – sadly, now it just recommends lawnmowers I should by.

  3. Very cool – super nerd brain on overdrive. Helps me, with my math disability 🙂 But I’m gratified to see your ability to translate and apply “class” work & theory to RL. Excellent 🙂

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s