09 Jan 09 Twitter social network leaders: navel-gazing or more?

I’m exploring the Twitter data I’ve gathered over the last few weeks, which is designed to uncover patterns of URL citations, which I believe is one of the service’s most powerful uses. As I have written, I’m looking at Twitter as a massively parallel self-organizing point-of-view system. In other words, my premise is that by posting URLs to Twitter, people are saying that they found a web page to be interesting and valuable.

Today, I’m looking at “centrality,” a typical social network metric. I am interested in degree centrality, which looks at how many connections a person has, which shows who the key players are. I’m considering two people to be connected if they cited the same URL in the same time frame, regardless of whether or not one was an explicit retweet of the other. Later, I’ll probably weight the connections with explicit retweet and other data. For now, I want to see if follower count, a far simpler metric than centrality, would work just as well. Here is a log-log scatterplot of degree centrality v. follower count.

Follower count v. degree centrality

The data points are scattered all over the place, which means that follower count does not correlate to the connections revealed by citing the same URLs. I’m not surprised, given all the games people play to get followers, the robots and such that have little or any human thought behind them.

As a reality check, let’s look at a similar plot that compares follower count to user mentions. I would expect that people who have a lot of followers will be mentioned (in the form of @screen name, in a reply, retweet or any other context) more often. Here’s the graph.

Followers v. mentions

Bear in mind that my data gatherer is biased toward people who cite a lot of URLs, so when I say count mentions, those are mentions by people who tend to cite a lot of URLs in their posts. As you can see, although there are many outliers, there is an obvious trend upward and to the right, which indicates a positive correlation – people with a lot of followers indeed do tend to be mentioned a lot. The upper left area is almost empty because it is hard to get any mentions when you don’t have any followers. On the other hand, you can have lots of followers and few mentions, which is why the there are more points toward the lower right.

Outliers are often interesting and I find myself wondering who is getting a lot of mentions even though they have very few followers. The dot closest to the upper left corner is MsTweet, who is a “customer service evangelist for Mr.Tweet” and therefore doesn’t follow much of anyone, but gets mentioned a lot. In the upper right border area, with lots of followers and mentions, are Shorty Awards, Chris Brogan, Guy Kawasaki, and ReTweetTrends (in the center of the top, not following nearly as many as the others). The lower right corner outliers are people who are heavily followed, but rarely mentioned by people who cite URLs. They include Kevin Rose, Jason Calacanis, Veronica and iJustine. I’m surprised, actually, that these folks’ huge followings apparently either aren’t mentioning them often or aren’t often citing URLs. Let’s reality-check that with Twitter search.

I’ll search on each of their user names, then repeat the search with their name and “http,” which will give a rough comparison of all mentions v. mentions with URLs in them. Twitter’s search doesn’t give a result count, so it’s pretty hard to tell. All I can go by is the frequency of recent tweets. Let’s compare it to somebody who is mentioned a lot – Chris Brogan. He is definitely getting a lot more frequent mentions in conjunction with URLs, so at first glance, the data seems believable.

Perhaps this indicates that the people with big followings yet few mentions have a different kind of influence. People like Chris and Guy seem to be leading others to look outside of Twitter, while Kevin, Jason, Veronica and Justine have some other, perhaps more Twitter-centric influence. Is it safe to say that the latter group is more engaged with Twitter for its own sake?

It seems that some of the popular Twitterers are leading their followers mostly into Twitter navel-gazing, while others are leading people beyond what Twitter itself has to offer. I find myself wondering how this might change as Twitter matures… and wondering if perhaps the navel-gazers are newer to Twitter and will get bored faster. I’m gathering more of the user information now, so I should be able to compare the average number of days they have been using it. In any event, from a business standpoint, I think I know which kind of leader I’d be more interested in.

Tags: social network analysis, twitter

Filed in social network analysis with Comments

01 Jan 09 Implicit social networks: If Guy Kawasaki is right, so am I

Just read Guy Kawasaki’s Looking for Mr. Goodtweet: How to Pick Up Followers on Twitter, in which he offered the following tip:

Tip 4: Follow everyone who follows you. When I first started on Twitter, Robert Scoble told me to follow everyone who followed me. “But why, Robert, would I follow everyone like that?” The answer is that it’s courteous to do so and because when you do, some people will respond to you and eveyone who follows them will see this—which is more exposure for you.

Having said this, when you get to more than fifty or so followers, it’s impossible to read what all your followers tweet. At that point, you have to focus on direct private messages (“Ds”) and direct public messages (“@s””).

Yipes.

The first analysis I did on Twitter was to count the followers of followers, as an illustration of how influence is a more-than-first-order phenomenon. The number of followers your followers have almost surely correlates to your potential influence. People like Guy are just about impossible to measure that way – they have so many followers that it is impractical to count their followers’ followers.

When I first looked at Twitterholic and saw how many followers Guy has, I thought “How the heck does anybody follow that many people?” That question is answered – he doesn’t. It is possible that Guy decided to do me a favor (along with everybody else who looks at implicit, in addition to explicit, social networks). Guy knows that this is where I’ve focused for years… but, okay, that’s probably not why he advises people (who want a lot of followers) to follow everybody who follows them. Still, it is good news for me.

The pattern of followers is Twitter’s explicit social network. As soon as I started analyzing it I was stymied a bit by robotic auto-followers. They play havoc with metrics that depend on the follower-followee relationship. My friend Dave suggested that Twitter might need user-agent meta data, which would reveal whether or not a given Twitter user is a real person or not. This would allow software to omit users like hashtags, twemes, AmazonGoldDeals, mrtweet and so forth, all of which automatically follow you if you follow them. But it wouldn’t solve the problem as long as there are people like Guy with robot-like behavior, automatically following everybody who follows them. I’m fairly certain that Guy really is not a robot – even though, in addition to tip No. 4, he advocates somewhat mindless direct replies:

Tip 2: Send @ messages to the smores. They probably won’t answer you, but that’s okay. All you want to do is appear like you have a relationship with them to enhance your credibility. The theory is, “If she is tweeting with @scobleizeer, she must be worth following.” Bull shiitake logic, admittedly, but it helps. To bastardize what a famous PR person once told me, “It’s not who you know. It’s who appears to know you.”

That tip guarantees that I’ll be sending Guy an @ message about the tweet that points to this post. But I actually do know him, unlike the 10 or 15 other people on Twitter who don’t, who will @ message him anyway, further confounding those who naively analyze the explicit social network by looking at @ message relationships.

There’s nothing wrong with analyzing the explicit social network, but it is a big mistake to trust the results by themselves. So I focus on the real meat in Twitter, figuring out as much as possible which items are getting real human energy into them and what they imply about relationships. When a hundred people post links to the same URL, odds are that the page they’re tweeting really is meaningful and those people are likely to influence each other in the Twitterverse, especially if they used the same “shrunken” URL. Throw in screen names, hash tags and language patterns and perhaps something truly useful and meaningful will come out. I hope so.

As long as Guy doesn’t start advocating retweeting the “smores” tweets, I’m probably okay. So far, all he’s done in that direction is to tell people to repeat their own tweets.

If Guy is right that it is a good idea to follow everybody who follows you, then I must be right to insist that any analysis that depends only on the explicit social network is inherently flawed. Now I just have to decide if I really want to follow, not just Guy’s advice, but everybody who follows me. A voice in my head is saying, “If everybody jumped off a cliff…”

Tags: Influence, social network analysis, twitter

Filed in Influence, social network analysis with Comments

07 Dec 08 More friends means more Twittering

Jeremiah tweeted about an HP Labs paper that show that the more friends and followers an active Twitter user has, the more they’ll post. The number of friends was more significant than the number of followers. I instantly found myself wondering if the numbers would correlate as well if the number of followers was not visible to users. This goes back to what I wrote the other day about treating a web analytics data warehouse as part of the production system. “Number of followers” is a simple metric, but it is a kind of feedback that isn’t so easily available in other contexts. On the web, counting unique visitors is among the most wretched of web metrics; counting unique RSS subscribers is muddied by aggregation.

The HP paper concludes that most of the relationships identified in a social network are weak; the strong social network, the real friendships, are hidden within it. Therefore, the authors argue, if you’re going to do viral marketing, you have to discover and tap into that hidden, more deeply connected network. I think they went too far there.

They assume that only the strongest friendships can mediate viral ideas. Why?. I don’t think the study really addresses that question. If you want to see the flow of influence through the network, you can’t just look at who is communicating with whom, you have to look at how people and network respond to communications.

A former intelligence guy explained it to me with a Cold War example. If you see a pattern that when radio station X transmits, a large part of the Soviet Navy suddenly changes direction, you can guess that X is a command and control center. That’s traffic analysis at its simplest.

In a network like Twitter, the equivalent would be to observe a correlation between person X’s posts and some measure of network activation, particularly something like “re-tweeting,” which has little cause-and-effect ambiguity.

Tags: social network analysis, traffic analysis, twitter

Filed in Uncategorized with Comments

Recent Posts

Recent Comments

Twitter: Web Analytics (#wa)

Blogroll

Tags

Somewhat gratuitous advertising