Under the Influence of...Twitter

Interesting post by Ethan Bauley on HP Labs' research on Influence on Twitter - confims something our own analysis was telling us:

According to the research, it is important to separate the concept of “influence” from “popularity.” While a user on Twitter may have a large number of followers, his or her influence is more strongly associated with their engagement with the network, rather than the raw number of followers or retweets.

To automatically identify influencers, the authors devised an algorithm called the IP Algorithm. This algorithm assigns a relative influence score and passivity score to every user:

- “Passivity” is a measure of how difficult it is for other users to influence him or her

- “Influence” depends on both the quantity and quality of the user's audience

The paper concludes: “This study shows that the correlation between popularity and influence is weaker than it might be expected. This is a reflection of the fact that for information to propagate in a network, individuals need to forward it to the other members, thus having to actively engage rather than passively read it and cease to act on it.”

As we remarked about a year ago, our initial analysis showed that the then-popular metric used by all the PR agencies of equating "Influence" and "Social Capital" on Twitter with "Followers" was miidly correlated (ie lame) at best.

The actual HP paper is over here and makes for interesting reading. I'd love to see their algorithms

Twitter is becoming quite the Digital Anthropology testing ground, the Digital Rwanda, as we noted a while ago on Twitterers in the Mist dealing with recent research on monitoring Twitterer happiness.

Also, at our TEDxTuttle 2 session we had the fascinating Dr Caroline Wiertz of Cass Business School giving us a sneak preview of their work looking at how Twitter influences consumer reaction to things like movie selection (especially opening weekend movies - Movie Critics beware).

It's interestng to think about why Twitter is such a good research ground. Our own experience is that it:

(i) Is very easy to parse as it is all easy to scrape
(ii) The tiny message size is (relatively) easy to analyse for meaning
(ii) The linkages are very easy to analyse and map so it really lends itself to mathematica analysis
(iii)There is a a large heterogenous population which is growing so it is increasingly representative
(iv)Topics covered are vast, from the sublime to the ridiculous via the intelligent and the lame.

I predict a plethora of PhD's in Digital Anthropology using Twitter. I do recall a piece of work done some years ago on predicting infidelity from telephone call patterns, so not all of this new analysis is going to be comfortable reading.

You have been warned

Update - by the way, my Twitter Inbox is deluged by people I follow who are p*ssed off with the new Twitter "Who to Follow" function. Could be that Twitter needs to get HP Labs and Dr Wiertz in there fast!