twInfluence uses social network analysis techniques to analyze your Twitter followers.
Here's the metrics they are using. I would like to get a sense of what they mean - perhaps scales? How to put this into context? I'm particularly interested in the social capital measure.
Here's the description from the site of the various metrics, but would really like to get a sense of how to use the numbers to improve Twitter presence - and what the scale/range is for some.
Definitions
First and Second Order Networks: From the perspective of graph theory,
a Twitterer's followers would be considered their first-order network,
and their "followers count" the same as their "degree". "Degree" is a
simple form of centrality measurement that equates to "prestige" or
"popularitiy"; different types of centrality can measure connectivity,
authority, and control in a network. The following diagram demonstrates
the different "neighborhoods" in a network. The Twitterer is the
primary node (shown in red); its first-order neighbors (shown in green)
surround it, and its second-order neighbors (shown in blue) surround
the outside.
Reach: Reach is the number of followers a Twitterer has (first-order
followers), plus all of their followers (second-order followers). In
the diagram above, the reach would be 27 (there are 28 nodes, including
the Twitterer). This is by necessity a crude maximum estimate, since
there will definitely be duplicates and overlaps that could only be
eliminated by up to thousands of API calls. Reach is a measurement of
potential audience and listeners, a best estimate of the number of
people that a given Twitterer could quickly get a message to.
Velocity: Velocity merely averages the number of first- and
second-order followers attracted per day since the Twitterer first
established their account. The larger the number is, the faster that
Twitterer has accumulated their influence. Of course, this number could
jump significantly with the addition of a few high-profile followers.
Social Capital: OK, I'm abusing the academic term "social capital" a
little to indicate the average first-order network of a Twitterer's
followers. A high value indicates that most of that Twitterer's
followers have a lot of followers themselves.
Centralization: This is a measure of how much a Twitterer's influence
(reach) is invested in a small number of followers. Centralization
scores range from 0% (completely decentralized) to a theoretical 100%
(completely dependent on one Twitterer). In social network analysis, a
high centralization indicates dependency of the network on just a few
nodes to maintain the connectivity of the entire network. Twitterers
with low centrality networks would not have their reach greatly reduced
if a few high-profile people stopped following them.
Efficiency: The more people you follow, the more time you have to spend
reading and filtering tweets. I've heard it argued that nobody can
effectively and consistently follow more than a couple hundred
Twitterers and actually keep up with the tweetstream. Efficiency is
simply my measure of how many people a Twitterer had to follow in order
to build up their reach, as a percentage. Highly "efficient" people
like Will Wheaton follow only a few dozen people even though they
themselves have thousands of followers! Ideally, this would measure a
Twitterer's Friends' collective "number of updates" relative to reach,
but we have to make an excessive number of API calls as it is.
I appreciate your explanation. I entered my own Twitter name into this website and wasn't sure how to interpret the results.
Posted by: Liz | October 14, 2008 at 07:05 AM
The explanation comes from the site - I want to know how to put the numbers in context - like for example the centralization -- what is the range of numbers? What is high versus low centralization number?
Posted by: Beth Kanter | October 14, 2008 at 07:53 AM
First, Beth - MAZEL TOV! Does "most reach" mean within your own network, or across Twitter? Either way, that's a great stat. I'd like to suggest we consider another layer in our analysis, though it's not reflected in the Twinfluence data: The content of the post. "What I'm doing" is very different than a post with a valuable nugget (like a link) and the way the tweet is composed ("great summary of... link.." vs. just "link"). The network and reach is necessary "infrastructure" to carry a message, but not sufficient for that message to actually saturate your first order network or move to the second order network. Though I have no idea how to measure this... thanks for your work on stats and measurement -- very much needed in our nonprofit field to get funding and support for social media work.
Posted by: Lisa Colton | October 14, 2008 at 08:17 AM
Beth, from what I understand of SNA, this is highly context dependent. So high centrality needs to be figured in comparison to others. Your centralization value suggests to me that you do not have a network all following a few core "famous people" but instead have a rich ecology. The difference between network as pyramid and network as starfish, if those references mean anything to you.
Efficiency is a bang for the buck. You follow Obama, and he has a zillion followers. So if you post something, and he reposts it, then you have a great efficiency in broadcasting messages.
Velocity is a growth figure. It is not of much value by itself, in my mind, as it is over time...who is bringing in new followers to your 2nd order network over time and at a high rate? Is that something you value or not? (I wouldn't pay any attention to it, but a marketing person might.)
Reach is sort of bizarre here...I mean we have all seen tweets get retweeted more than once. Your actual reach on twitter is likely higher than just first+second order.
I am a bit offended by Social Capital being measured by network size. Social Capital has a lot more to do, in my estimation, with the quality of connection. How to measure social capital depends on what is valuable within that social network but is, I believe, related to *trust* and not size.
I hope that helps. Valdis Krebs was my introduction to SNA, and his white papers on orgnet.com are fairly approachable/digestible. See also networkingweaving blog. You might really enjoy learning what June Holley has been working on, if you are not already familiar.
Posted by: Jean Russell | October 14, 2008 at 08:35 AM
I'm with Jean Russell on feeling that this is a gross simplification of the meaning of social capital (and you did apologize for taking liberties with the academic meaning). I'm a grad student studying it and won't bore you with the analysis here, but it goes well beyond quantity to the quality of the connections.
One of the things missing here that existed in some of the sociological research on social networks is the idea of valence of the connection. That is, I may KNOW you, but are my feelings toward you positive or negative? If you could assign a +/- to the relationships in Twitter, that would affect the calculation of someone's influence.
This is particularly relevant during a political season. I doubt everyone following either of the presidential campaigns is a supporter--there are plenty of people doing opposition research. I even see that at my local level, where a campaign supporter of an incumbent county commissioner is following the opponent on Twitter.
There is another concept--can't recall the sociology term--that has to do with the fact that relationships are not always equal. The influence doesn't necessarily flow in both directions at the same rate or with the same effects. So I have a connection with my boss, but I don't tell him what to do. This gets at the influencer concept so inherent in social networks that I think they're trying to measure. But influence is also more than quantity.
A lot is lost in the averaging function. I have very few followers (20 as of today). One of those, however, is Chris Brogan. So my second-order numbers are incredible, mostly thanks to him (thanks Chris! For that matter, thanks for leading me to Beth's Blog).
But I don't kid myself that many of my posts are going to be retweeted, given how many followers he has. The efficiency issue must be very real for him (which is probably why he's using TweetDeck to organize). (Lisa's comment about the content of the post is also relevant--I really don't expect Chris to tell everyone that we took our kids apple picking yesterday).
Just a bit of a weakness in the oversimplifications of the data. It still has some value--it just won't tell you how many people you're REALLY reaching.
@BarbChamberlain
Posted by: Barb Chamberlain | October 14, 2008 at 09:18 AM
I personally love this, if for no other reason than it can serve as an example for people who may not "get" Twitter. Even if the data is not super specific or even relevant--the the reach category in particular has a good, visual "ah ha" factor.
Posted by: Maggie McGary | October 14, 2008 at 10:14 AM
I forgot to ask: is there an equivalent of this for Facebook?
Posted by: Maggie McGary | October 14, 2008 at 10:45 AM
Maggie - There are some gizmos like this for Facebook, like the Six Degrees or Friend Wheel applications. Don't know if they are still live, though.
Beth - With respect to these measures and some sense of scale, you have to realize that you are off the scale in general. If you have so many people in your friends and friends-of-friends lists that you've hit Top 5, you are way high up there.
On the opposite end, my brother has 4 friends and 5 followers. He's got a reach of 9,444; velocity of 56/day; social capital of 1888.8; centralization of 59.54%; efficiency of 2361 (which is 9444/4).
And me, likely on the higher end, though I don't know. 243 friends, 404 followers. Reach 418,082. Velocity 924/day. Social Capital 1034.9. Centralization 8.73%. Efficiency 1720.5.
And what do these number mean? Reach is simply "how many people are likely to see something I write." Via my followers or their followers. Social Capital is similar, though it is probably a "how fast" will that message get out (Reach / Followers). If it's high, then your followers have lots of followers too. In the case of my brother, he has one follower that tips the scales. Given Beth's reach, I'm thinking the "social capital" number is fairly meaningless.
Most of these are explicitly outward facing measures: how does my message get out. I'd like to see similar metrics on where I get my message from (other than number of friends).
There are some other metrics of value to SNA. What is the intersection between friends and followers? Is there a total overlap or very little overlap? Of the non-overlappers, how much "reach" is there in both directions?
Posted by: Jack Vinson | October 14, 2008 at 12:28 PM
Hm. I think they're fudging the figures in their 'top 5 lists' at the bottom. take a look at yours, and them mine. See http://www.downes.ca/post/46646
Posted by: Stephen Downes | October 14, 2008 at 03:15 PM