Twitter List Analysis? (Anyone know of interesting, useful, or competent analysis of lists on Twitter?)

Replies to This Discussion

Permalink Reply by Gerry Crispin on January 15, 2010 at 3:42pm

Assuming I had a list of people I want to influenvce or, be influenced by, AND they were on Twitter, then the list I would analyze is not the one they make up but the list of poeple they actually follow...which is available for inspection just not easily imported for analysis. In some instances we might be looking for and uncover a mobius strip effect (everyone flows each other) or, perhaps we might dig deeper and learn about "Case 0"- the one person all influencers follow.

▶ Reply to This

Permalink Reply by Irina Shamaeva on January 15, 2010 at 3:57pm

Very interesting thoughts, Gerry! How does one influence others, and what role can Twitter lists play?

There is a recently published list of 25 Most Influential Recruiters on Twitter

I haven't seen any good tools for twitter lists analysis. In fact, I prefer TweepML lists to twitter lists.

There's a great tool to analyze one's following, FriendOrFollow. It can be used to analyze lists (creat a new twitter ID and import your list to do that).

Irina
@braingain

Gerry Crispin said:

Assuming I had a list of people I want to influenvce or, be influenced by, AND they were on Twitter, then the list I would analyze is not the one they make up but the list of poeple they actually follow...which is available for inspection just not easily imported for analysis. In some instances we might be looking for and uncover a mobius strip effect (everyone flows each other) or, perhaps we might dig deeper and learn about "Case 0"- the one person all influencers follow.

▶ Reply to This

Permalink Reply by Slouch on January 15, 2010 at 4:11pm

The problem with that list Irina and of course there are problems with all lists of these kinds is that the #1 spot is Shally who I really do like but he is not even a recruiter. Maybe he influences recruiters but he himself is not.

Gerry, I have something Im working on called whyIfollowU or maybe you - still deciding and it would be why you follow someone else. that would have value or at least maybe a little more value.


Irina Shamaeva said:

Very interesting thoughts, Gerry! How does one influence others, and what role can Twitter lists play?

There is a recently published list of 25 Most Influential Recruiters on Twitter

I haven't seen any good tools for twitter lists analysis. In fact, I prefer TweepML lists to twitter lists.

There's a great tool to analyze one's following, FriendOrFollow. It can be used to analyze lists (creat a new twitter ID and import your list to do that).

Irina
@braingain

Gerry Crispin said:

Assuming I had a list of people I want to influenvce or, be influenced by, AND they were on Twitter, then the list I would analyze is not the one they make up but the list of poeple they actually follow...which is available for inspection just not easily imported for analysis. In some instances we might be looking for and uncover a mobius strip effect (everyone flows each other) or, perhaps we might dig deeper and learn about "Case 0"- the one person all influencers follow.

▶ Reply to This

Permalink Reply by Joshua Letourneau on January 15, 2010 at 4:47pm

Thanks for the thoughts, everyone.

I think, at this point, the idea is to simply measure the data points and investigate 'listing' phenomenon, such as:

Is there a clustering effect that follows 'listing'? Do 'listing' structures follow small-world phenomenon, such as what naturally occurs within human networks, protein networks, food chains, "2nd Life" social networks, etc.

How many degrees out does 'listing' spread? Does it increase at the 2nd degree and further at the 3rd degree, and then throttle back? Is there a single wave pattern, or is it more chaotic, such as a series of waves (i.e. a bunch of rocks hitting a pond at the same time, etc.)

If a friend lists you, how likely is his/her friends, and/or further, their friends?, etc.

There could be some interesting data that comes from this.

▶ Reply to This

Permalink Reply by Joshua Letourneau on January 15, 2010 at 6:00pm

We're not getting into qualitative analysis yet. We're talking quantitative analysis. Once we get the data, then we can speculate. Think of this like a science experiment. Right now, we're in the dark.

Please don't assume we're measuring "influence". We're trying to discover; walking down the aisle, not knowing what we'll find.

So right, we're an open book walking in with no assumptions. Let's see what the data says, and then we can evaluate.

▶ Reply to This

Permalink Reply by Joshua Letourneau on January 15, 2010 at 8:09pm

Honestly, I really don't know. I haven't been given the skinny, to be honest. It's just me doing a favor because there aren't too many 22k member communities of so many people on the cutting edge of technology. I don't even think Marketers are as 'technology-progressive' as we are because they're not constantly hunting. Farming, maybe . . . but hunting? That's for the sales folks, right? :)

Anyway, I wish I could figure out what kind of value something like this might bring. I think it's just intellectual curiosity if you ask me :)

Then again, maybe I'm thinking too small.?.

P.S. Quantitative research just means we're only looking at the data (in this case list numbers), but believe it or not, sometimes what we believe to be true is proved incorrect by the numbers. Qualitative just means you look at behavior, so it's more a 'social sciences' kind of thing. Thinking of 'Social Studies', that just took me back to grammar school. Jeez, I've gotten old :)

P.S.S. Karen, you know what could come out of it? Understanding of how something like 'listing' spreads. What can you do with that intelligence? Game the system of course! :) Just like you can game your SEO to achieve 1st page Google Ranking :) It's all a cat-and-mouse game.

▶ Reply to This

Permalink Reply by Irina Shamaeva on January 16, 2010 at 12:56pm

Josh,

Speaking of the twitter lists functionality, I am not sure I see the value of it other than expressing your views through lists and, more importantly, coming across as an expert: if someone (TwitterID X) creates a list Y about subject Z that probably means that X is an expert in subject Z.
It could be interesting to watch someone's lists's tweets - but I am not sure I know how this knowledge can spread. Perhaps I am missing something here. I would be interested in hearing from others.

If you create a list on TweepML, this is different: then others can follow each one of the people on the list and possibly make personal connections.

Apart from lists, the subject of influence and how it can be measured is also interesting.

Irina

▶ Reply to This

Permalink Reply by Joshua Letourneau on January 16, 2010 at 7:17pm

Hi, Irina - yeah, you very well may be right. Perhaps he doesn't find anything interesting. Sometimes an experiment doesn't work out in terms of showing us anything special :)

We're on a slippery slope if we start throwing the term "influence" around lightly, so I hesitate measuring something like that using mathematical/quantitative means. However, if we use the precursor "digital" (i.e. "digital influence"), then we could define it any way we want and use quantitative means to measure it. An analogy is saying "sexual relations" instead of "sex". Per Bill Clinton, the single word "relations" completely changes the definition :) I'll pass on that, though. Influence means "to create change", so I really don't know how we can quantitatively measure something so subjective in the given context of this question.

Irina Shamaeva said:

Josh,

Speaking of the twitter lists functionality, I am not sure I see the value of it other than expressing your views through lists and, more importantly, coming across as an expert: if someone (TwitterID X) creates a list Y about subject Z that probably means that X is an expert in subject Z.
It could be interesting to watch someone's lists's tweets - but I am not sure I know how this knowledge can spread. Perhaps I am missing something here. I would be interested in hearing from others.

If you create a list on TweepML, this is different: then others can follow each one of the people on the list and possibly make personal connections.

Apart from lists, the subject of influence and how it can be measured is also interesting.

Irina

▶ Reply to This

▶ Reply to This