Top thinkers (and doers) in Canada’s #opengov (open government) space are meeting in Ottawa for the Open Dialogue Forum March 31st and April 1st, 2016. Not surprisingly, given the audience, the Twitter chat was on point. I’ve done a quick analysis of the mention network from the #CODF16 tag in order to map out who is talking and who is being heard.
For those not familiar with network graphs, the dots represent Twitter accounts and lines represent when one account mentions another. In network science those dots are called nodes and lines are called edges.
This graph shows everyone who made a tweet using the #CODF16 hashtag on March 31, 2016. The colours represent clusters of account who tend to mention each other. The nodes are bigger the more times someone else mentioned them. Obviously this graph is pretty messy and hard to understand so lets break it down.
Top level (the tl;dr you busy folks):
- There is one major community of accounts chatting but sub-communities (“clusters”) exist.
- There are no clear divides based on type of stakeholder which means conversation is happening among different kinds of actors (a key goal of the forum).
- There is an imbalance in who mentions and who gets mentioned which probably means the Twitter conversation is not reaching all types of stakeholders (notably, politicians and journalists seem to be largely absent from discussion).
I’ll talk to you, blue:
The biggest cluster of accounts is light blue. That big node in the middle is@CdnOpenDialogue (the event’s official account). I know, what a surprise, the ones who organize it are mentioned a bunch. In fact that account was mentioned 118 times (they mentioned others 65 times).
But, there is another account that was mentioned even more: MP Scott Brison. He got 130 mentions while only sending out 2 himself. He is in the dark blue cluster:
Where the conversation is happening:
These two examples are pretty typical of Twitter chat. We see accounts with some type of popularity have high “indegree centrality” — that means they are mentioned a lot. But, what about those doing the mentioning? They are normally the ones asking questions, responding to comments and sharing different ideas. We can find them by looking for “outdegree centrality” — those who mention a lot.
Let’s look at that first cluster again:
We’ve still got the organizers but now another node pops up.@DoreySamson was mentioned 14 times but made 64 mentions.
Similarly, in another cluster Canada 2020 (on Twitter) pops up:
And, you guessed it, @ScottBrison is no longer particularly central, but others around him are:
When you dig deeper into the clusters you see a lot of different kinds of stakeholders in each one. Think tanks, government officials, academics etc. are all communicating with each others. That said, those who are most engaged in back and forth communication tend not to be the organizers or politicians.
A word about words:
I can’t do the content justice in this quick post so I won’t provide analysis of what people are saying at this point. But I do want to share a few quick things:
- “Open” is the most frequently used word in the dataset.
- #OpenGov is the hashtag most often used along side #codf16.
- There were 2894 tweets made by 497 unique accounts at the time of analysis.
What it means:
Open Dialogue Forum has so far succeeded in generating chat among a variety of stakeholders. For the most part these various types of stakeholders are engaging with each other. Notably, there are not many journalists engaged in this network which is pretty abnormal for any kind of political chat in the Canadian Twittersphere.
I haven’t been studying this community over time so I can’t say if the forum was the cause of these connections but I can say that it has done a good job of sparking conversation.