Freakonomics has more than 270,000 followers on Twitter. And we love every one of you. No, really! But we don’t follow anyone back. (When we first signed up with Twitter, some very smart media consultants said that reciprocity was the name of the game; but we didn’t want to get involved in some massive online tit-for-tat.) So, does that make us
That’s one of the questions we ask in our latest podcast, “Is Twitter a Two-Way Street?” (You can download/subscribe at iTunes, get the RSS feed, listen live via the link in box at right, or read the transcript here.) We try to find out just how social you need to be on a social-media site; if you want a lot of followers on Twitter, do you need to follow a lot of other Tweeps?
You’ll hear from the network-theory whiz Duncan Watts, a former Columbia sociologist who now works at Yahoo! Research. He’s the author of the very good book Six Degrees, as well as the brand-new Everything is Obvious (Once You Know the Answer). And he co-authored a recent paper called “Who Says What to Whom on Twitter.” From the podcast:
WATTS: It’s true that, you know, there are millions of users on Twitter who are listening to other users. But we also find that there is a remarkable concentration of attention. So about 50 percent of all tweets that a random person on Twitter receives on any given day come from just 20,000 users. So that’s about .05 percent of…so one half of a tenth of a percent of all users on Twitter.
WATTS: It’s worth emphasizing again that Twitter is not a social network. Now, social networks are characterized by very, very high levels of reciprocity. So if I say that I’m friends with you, it’s very likely that you will also say that you’re friends with me. It’s not always true, but it’s very often the case. And if not, I stop being a participant in that social network. It’s a funny kind of friendship if only one person thinks that it exists. Whereas, in communication networks it’s totally different.
You’ll also hear from Justin Halpern, who parleyed his Twitter feed “Sh*t My Dad Says” into a best-selling book and a TV show. Halpern has millions of followers but follows only one other person. (I don’t want to ruin the surprise by naming the person here.)
And you’ll also hear from Steve Levitt, who talks about a fake Steve Levitt Twitter feed and tries to help me come up with a Freakonomics following strategy. For now, we’ve latched onto one person.
Anyone who cares about Twitter should spend some time on Twitaholic, a site that scrapes follower/followed data from Twitter. You’ll see that the top Tweeters — Lady Gaga, Justin Bieber, Britney Spears, and Barack Obama — have a relatively high follower-to-follow rate. Lady Gaga has nearly 8.7 million followers, and follows more than 144,000 users; Obama has nearly 6.9 million follows, and follows more than 700,000 users. (Presumably someone other than the President himself does the following.)
All of which seems to argue that reciprocity matters a great deal. But once you look below the very top of the list, the ratio plunges. So: how important is reciprocity?
We asked Duncan Watts to look into the follower/followed numbers and see what trends he could identify. He came back with some results that he warns are “quick-and-dirty,” but worthwhile nonetheless:
Here are two sets of results (1) for the top 1000 most-followed users; and (2) the 1000 users who follow the most others (i.e. with the most “friends” in Twitter parlance). [These numbers were taken from Twitter a few weeks ago, so are slightly outdated]
In both cases, the x-axis is # friends (i.e., how many they follow) and the y-axis is the # followers. The results are binned, so the labels are a bit deceiving, but you get the drift. The boxes represent the interquartile range; the horizontal lines in the boxed represent the median; and the bars represent the 95% confidence interval.
1000 Most-Followed Users:
Bottom line is that there isn’t any trend: for the vast majority of most-followed users, there is no relationship between following more people and being followed by more people. The only exception is the far-right bin, which actually contains just a handful of users, listed here:
And for these users, it’s not clear what is going on. BarackObama may follow lots of people, but it’s unlikely that that’s why he has lots of followers. Same for Britney or Yoko Ono. Guy Kawasaki and StartupPro are more interesting: they have almost exactly as many followers as friends. My guess is that in these cases, the causality is the opposite of what you were told — that is, they automatically follow anyone who follows them, so it is followers leading to following, not the other way around.
1000 Users Who Follow the Most:
Here we do see a clear positive trend: on average, the more users they follow, the more followers they have. But again, it’s not clear what the explanation is. Some of them are probably spammers, and in those instances, it probably is the case that they are getting followers by following tons of people — as there wouldn’t be any other reason for people to know about them. But other instances may be a result of well-known people auto-following people who follow them. That would also generate a positive trend, but the causality would be the opposite. Without looking at the data over time, it would be impossible to differentiate these two explanations, and even then it would be hard.
Thanks especially to Duncan for all his input on this episode. Feel free to tweet it.
Is Twitter a Two-Way Street?
Stephen J. DUBNER: “So, Levitt, when we first thought about starting a Twitter account for Freakonomics, a couple of smart media consultants told me that it would be very poor form to expect people to follow us unless we followed a lot of people as well. There’s really a reciprocity at work here. And we didn’t follow that advice. We follow zero people on our Twitter account. Um how do you feel about that?
Steven D. LEVITT: We have a Twitter account?
ANNOUNCER: From American Public Media and WNYC, this is Freakonomics Radio. Today, how anti-social can you be on a social networking site? Here's your host, Stephen Dubner.
DUBNER: So we did start a Twitter account. But we aren’t what you’d call aggressive Tweeters. We don’t tell people what we had for breakfast, or what show we’re watching on TV or which kid lost which tooth. In fact, all we really do is send out links to our blog posts or this podcast, stuff like that. But here’s the thing: we have a lot of followers -- at least what seems to me like a lot of followers -- about 250,000 people. Now, that probably doesn’t mean much. It costs nothing to follow someone on Twitter; all you have to do is click your mouse one time. A lot of those people probably never read a single thing we tweet. But still, it’s kind of cool …
But as I told Levitt, we don’t follow anyone on Twitter. It just seemed like, if you’re going to follow some people then you’d feel bad about not following other people and next thing you know, you’re spending your whole day on Twitter, figuring out who to follow and who to be followed by.
So for us, Twitter is a one-way street. It’s a little bullhorn, nothing more.
So here’s a question for you: does that make us jerks? Or, since we’re talking about Twitter, does that make us twerks? Do you think Freakonomics should start following people on Twitter?
Duncan WATTS: OK, my name is Duncan Watts, I’m a principal research scientist at Yahoo! Research where I run a group called the human social dynamics group. And we’re interested in all sorts of questions that have to do with social networks and how information diffuses through social networks, how people influence each other, and how all of this helps us to understand social behavior.
DUBNER: Duncan Watts is a sociologist who taught at Columbia before moving on to Yahoo! He’s at the forefront of what’s called “network theory” -- how people are connected, whether in person or virtually, and what those connections yield. He’s written a few books -- Six Degrees, a sort of academic take on the Kevin Bacon thing; and a new book, Everything is Obvious -- Once You Know the Answer. It’s about “common sense,” and how it lets us down. Watts still writes a lot of academic papers too. His latest is called “Who Says What to Whom on Twitter.”
WATTS: It’s true that, you know, there are millions of users on Twitter who are listening to millions of other users. But we also find that there is a remarkable concentration of attention. So about 50 percent of all tweets that a random person on Twitter receives on any given day come from just 20,000 users. So that’s about so one half of one-tenth of a percent of all users on Twitter.
DUBNER: What do you call this as a sociologist, then, in terms of the distribution?
WATTS: Well, it’s a skewed distribution. But you certainly see this kind of distribution in activity if you look at, you know, how active people are on Twitter you see the same thing where there's a small number of people who are very, very active.
DUBNER: Were you surprised to find a concentration that intense?
WATTS: Well, we are used to seeing these skewed distributions so I think not in principle. It was still striking just how concentrated it was.
DUBNER: It may be more striking to people who don’t know what these distributions usually look like. I mean it may be more surprising to people who’ve been hearing for the past couple years that Twitter is the great democratization of communication.
WATTS: And it is, but what happens in democracies is that everybody pays attention to the same people. You know, so I think that it might change our view of democratization.
DUBNER: So, a relatively tiny group of people on Twitter wield most of the power. Remind you of anyplace else you know? Like, the offline world? Duncan Watts says he became a sociologist to study exactly this kind of thing: whose voices get heard in social situations, how people in groups interact, how groups form, how firms form, how markets form. This is the kind of thing sociologists have been fascinated with since the beginning of, well, since the beginning of sociology. They call it the micro-macro problem. In other fields, it’s sometimes called “emergence.” It’s when you put a bunch of elements together and somehow come out with more than just the sum of its parts. Trouble is, for a sociologist like Watts, this kind of thing has been pretty tough to quantify. Or at least it used to be.
WATTS: The problem is that actually measuring any of this, observing any of this has been historically impossible. So although we have theories about social networks that go back 50 or 60 years, and the sort of quantitative study of social networks goes back almost as long, in practice it’s been restricted to very small groups of people…
DUBNER: As many people as you could hit with a clipboard and questionnaire, right?
WATTS: Something like that. You’re asking…you know, you’re handing out survey tools, or in some great, some of the classic studies sociologists would even sit in a doughnut shop and record painstakingly every single time a person talked to another person, and they would sort of extract the communication network out of these interactions they observed, which is very creative.
DUBNER: Which, today…But today seems extremely archaic, right?
WATTS: Well it is.
DUBNER: The sample size is tiny, the sample pollution is strong, I guess, depending on which coffee shop you happen to pick, right?
WATTS: And you can only do it until your brain explodes, which is for must humans is a couple of hours. So, you can’t really, sort of, measure anything or observe anything that’s happening over extended periods of time.
DUBNER: The mountains of data being generated in an online ecosystem like Twitter are enough to make a sociologist like Duncan Watts put down his clipboard and drool. Twitter has about 200 million registered users, sending out more than 130 million tweets -- that’s 130 million data points -- every day. Even from those broad numbers, you can tell -- for every aggressive user, tweeting let’s say 20 times a day, there’s an army of folks who just sit still, keep quiet. Or maybe who signed up just because everybody else was signing up, the way everybody else used to sign up to write a blog, and then abandoned it. But if you’re a sociologist, even these things are good to know! Social media sites like Twitter and Facebook are changing the way academics see the world. And they may also change the way people like you and me see it.
WATTS: What we’re seeing is actually not different from how people behave offline, it’s just that we have a vastly increased ability to observe it, so it sort of seems different. People seem to think that they have many more friends now because of Facebook than they used to have. And that at the same time the quality of those friendships is somehow diminished. I’m oversimplifying, but this is sort of a refrain you hear over and over again. Particularly in the media, people are sort of wringing their hands over how friendship has become somehow diluted. But for most people this is actually not true. Right, for most people, this is actually not true. For most ordinary Facebook users, the people that they’re friends with on Facebook are in fact people that they know. Right, not through Facebook, but through some other means – because they went to school with them, or they work with them, or they met them at a conference or they had some interaction with them at a social gathering. Now, many of them may actually not be close friends, and without Facebook around they may not have a record of these interactions existing. So if you asked them in a pre-Facebook world how many friends they had, they probably would say, “Oh a few dozen or something,” because they would just be thinking about people who they really count as friends. But now we have Facebook to remind us that we have all these, sort of, vastly larger halo of peripheral relationships so we sort of feel like we have more friends and somehow that they’re less real than the ones that we used to have. But actually, we always had them. So there’s sort of an interesting kind of measurement effect here where you simply allow people to measure things and it changes their perception of those things. So it’s sort of, you know, I don’t want to say that nothing is different on online, because clearly there are things that are different and it may just be because it’s anonymity. If you took the anonymity away from online conversations, if you made people disclose, you know, if you made them log in through Facebook or something where they had to disclose their true identity, I bet a lot of that would go away.
DUBNER: Anonymity is strong. I mean you’re right, the division or the gap can be even something less profound than anonymity. If you’re driving in your car and somebody cuts you off, you might flip them the bird or do what. But walking on the street and they cut you off on the sidewalk, the physical proximity changes everything.
WATTS: Right, that’s a great analogy. In fact, there’s a, I’m sure you’re familiar with the classic obedience studies of Stanley Milgram back in the 1950s, and he found exactly this kind of result. The subject of the experiment was told that he was conducting a learning experiment on someone else who turned out to be an actor, and he was supposed to be giving this person electric shocks whenever they made a mistake. So the actor was sort of pretending to be, you know getting more and more tortured by these shocks, and the shocking result was that a remarkable number of people cranked up the voltage to sort of lethal levels simply because some experimenter was telling them to do that.
Now, what a lesser-known result of those experiments is that Milgram tried a bunch of different conditions. You know, in one case they actually had to sit there and hold the subject’s hand on the plate. So they were sort of physically in contact with the person they were shocking. In another one, the guy was visible but in another room. In another one he was in, on the other side of a wall so you could hear him, but you couldn’t see him. And sure enough, the further the person was away, the more socially distant they were, the more inclined people were to exercise what seemed like cruelty.
DUBNER: I didn’t know that. That’s interesting.
WATTS: So I think that’s an excellent point that you raise.
DUBNER: Coming up … what do Barack Obama and Lady Gaga have in common -- and why is Justin Halpern way cooler than either of them?
ANNOUNCER: From American Public Media and WNYC, this is Freakonomics Radio. Here's your host, Stephen Dubner.
DUBNER: There’s a website called Twitaholic. It tracks the most popular Tweeters -- shows how many people follow them, and how manythey, in turn, follow. If you like Twitter at all, you have to go to this site. You can thank me later. At the top of the list is Lady Gaga with nearly 8.5 million followers. Next are Justin Bieber, Britney Spears, and Barack Obama. The President has about 6.8 million followers. And here’s what’s interesting: all four of these people also follow a lot of Tweeters, at least 100,000 each. President Obama -- or, more likely, someone who works Obama -- follows more than 700,000 people.
But No. 5 on the list is Kim Kardashian, with nearly 6.5 million followers. And you know how many people she follows? 118! Al Gore has 2.2 million followers; he follows 9 people. In fact, once you get past that top 4 of power Tweeters -- the Gaga-Bieber-Britney-Obama quartet -- the ratio plummets.
Which made me wonder: if you want a lot of Twitter followers, do you need to follow a lot of people yourself?
I asked Duncan Watts to look into the numbers for the top 1000 users. His conclusion: there’s no trend, no correlation between following and being followed. But still … if our online lives really are just an extension of our offline lives … just as a matter of common courtesy …shouldn’t you reciprocate?
HALPERN: Uh, my name is Justin Halpern, and I created “Shit My Dad Says” and I’m author of the book by the same name and one of the writers of the television show by the same name.
DUBNER: Alright, and Justin how would you then assess the importance of Twitter in your life and career?
HALPERN: I would say it is possibly the most important thing aside from my father. Uh, without Twitter I definitely don’t think any of what has just happened in my life happens.
[EXCERPT FROM TV SHOW]
DUBNER: Here’s the blurb from Halpern’s Twitter page, it says: “I'm 29. I live with my 74-year-old dad. He is awesome. I just write down shit that he says.” The shit that Halpern’s dad says has attracted a lot of readers -- more than 2 million followers on Twitter. In fact, it was his Twitter feed that led to the book that led to the TV show. He has more Twitter followers than Arnold Schwarzenegger, Lindsay Lohan, and the Dalai Lama.
DUBNER: So as of this moment, it looks like you have 2,016,224 people following you on Twitter. And what I want to know from you is with more than two million people following you, I mean, that’s a lot of people, how many do you follow?
HALPERN: I only follow one person.
DUBNER: Who do you follow?
HALPERN: Uh, I only follow LeVar Burton. Of “Reading Rainbow” fame.
DUBNER: “Reading Rainbow” fame. And also “Star Trek: The Next Generation” fame , “Roots” fame.
HALPERN: That’s true. I did not give him enough accolades.
DUBNER: And you thought, if I’m going to follow one person, LeVar Burton seems to be deserving of that honor?
HALPERN: Yeah, I did think that.
DUBNER: And uh, at the time that you decided to follow LeVar Burton and LeVar Burton only, how many followers did you have?
HALPER: I had zero.
DUBNER: Oh! So this was before “Shit My Dad Says” was “Shit My Dad Says” even.
HALPERN: It was. It was. This was when “Shit My Dad Says” was read by me and one friend who didn’t have a Twitter account.
DUBNER: So, forgive my ignorance on this score, but I see that you generally, at least in the last year let’s say, you haven’t tweeted very much. Maybe fifty tweets in the past year. Which, look, if you can build a brand called “Shit My Dad Says” out of shit your dad says and just do it in just fifty posts a year, that means that you are wonderfully efficient and economical, but back in the day, were you tweeting a lot more?
HALPERN: I was. When I first started, I was living with him and normally sitting next to him for like eight to ten hours a day working, so I was getting a lot of stuff, and I would tweet, like you know, one thing a day and then since I have been working and I haven’t been near him as much, it goes down.
DUBNER: So what does it say to you, that you, a guy who is tweeting primarily or maybe even only things that someone else, i.e. your father, actually says, and you are only doing it fifty times a year, and yet you still have two million people following you. What does that tell you about the dynamic or the bullhorn nature of Twitter?
HALPERN: I think if you like the character of my father, then you know that you are only going to get stuff that he says. And I think that’s worked really well. I don’t think people really cared if I interacted with them or not.
DUBNER: Have people contacted you though over the years and said, “Hey, you know, you’ve become a big deal guy now with ‘Shit My Dad Says’ and I follow you and I like it, but man, you only follow one guy and it’s another famous guy. That’s just not fair.” Do people give you trouble for that?
HALPERN: Yeah, I got one...one message was, "Who do you think you are to only follow one person?" Uh, and I didn’t really have a response to that, other than I don’t think I’m anybody, I just only follow LeVar Burton.
DUBNER: So Justin Halpern has had incredible success on Twitter, which is a social-media ecosystem, by essentially being anti-social. Is there anything wrong with that? If people like to follow him, who are we to say that he has to reciprocate? At least he’s not what Twitter insiders call "the one-night stand" -- where you sign up to follow lots of people, hoping they'll follow you back, then you dump them a day later. Here’s Duncan Watts again.
WATTS: Well, I think Justin Halpern might be more the exception than the rule. There are sort of bona fide Twitter-generated celebrities, people who were not known beforehand who became known through their activity on Twitter. Although, even Justin Halpern probably wouldn’t be nearly as famous as he is if he hadn’t got a book deal that became a best seller, and a TV show, so you’re always sort of…One of the dangers of studying a single platform like Twitter is that you see a signal on it, and you want to understand the cause -- why did somebody become popular -- and the answer often lies outside the system that you’re studying. So most of the, I think all of the top ten most followed people are household names, right, Ashton Kutcher, Oprah Winfrey, Lady Gaga. I mean these are people who were famous before people came along and they are still famous. And they’re famous not because of Twitter, but because they’re on TV all the time and they’re in all of the celebrity magazines and there’s a whole sort of much, much larger media ecosystem that is sort of constantly putting them in our faces.
DUBNER: But let me ask you this, if I look at the very top Tweeters, let me take the top four, we’ve got Lady Gaga, Justin Bieber, Britney Spears, and Barack Obama. So those are as you said household names to the nth degree. Now they have, the four top followers have between six or seven million followers each let’s say. But they also follow a lot of people. So, Obama for instance has six and a half or seven million followers, but Barack Obama follows more than seven hundred thousand people on Twitter. Now, we assume that he’s not actually reading their tweets, so what’s the point?
WATTS: Well so, again, it’s worth emphasizing again that Twitter is not a social network. Now, social networks are characterized by very, very high levels of reciprocity. So if I say that I’m friends with you, it’s very likely that you will also say that you’re friends with me. It’s not always true, but it’s very often the case.
DUBNER: And if not, then I stop being a participant in that social network.
WATTS: It’s a funny kind of friendship if only one person thinks that it exists. So whereas in communication networks it’s totally different. The entire nation can watch Barack Obama give the State of the Union address, but he can’t watch everybody’s YouTube videos.
DUBNER: True enough, but what would the purpose be then, if I’m Barack Obama and I have a Twitter feed, and I, or someone around me presumably, not me myself, has come to the conclusion that we should tweet to get out message out. It makes perfect sense and we should get millions of followers because we’re communicators almost above all. But also why do I want to follow seven hundred thousand people? What’s in it for me? Is it just the appearance of reciprocity that’s supposed to translate into some general feeling of goodwill?
WATTS: Well, I actually think that…my guess is that different kinds of users have different reasons for using Twitter.
DUBNER: Let me ask you this, there are some people then who are followed by a great, great. great many people and yet who follow nobody. So, for instance Stephen Colbert is followed by more than two million people, it’s a lot of Twitter followers, and follows zero. First of all do you have a name for people like that, and what can you say about them?
WATTS: Well, putting on my amateur armchair psychologist hat here, I would say that it’s almost a status symbol to be followed by many people and follow very few. It’s sort of like having lots of followers even though you don’t tweet very much. It’s sort of like, well I’m not even trying and I’m still popular. But I’m going to again guess that, you know, these are all individual people with their own agendas and psychologies, and you know, there’s probably as many reasons for these patterns as there are people.
DUBNER: Not even trying and still popular. Wow. Is that how we want to be? I’ve seen first-hand how successful Justin Halpern is, and he only follows Geordie La Forge. I heard Duncan Watts say that you don’t necessarily have to follow to be followed... Still, is that how Freakonomics should behave on Twitter? Steve Levitt and I had a summit the other day. We talked it over.
LEVITT: Well, I would say given that neither you nor I has ever gone on Twitter other than to send out our blog post, that, why don’t we follow everyone? Since we don’t look at what they are saying anyway, and if it makes them feel good to follow them, why not follow every single person on Twitter? That could be our claim to fame is that we follow every person on Twitter. As long as we never look at the account it won’t cost us anything.
DUBNER: I like it. I like the strategy. Or, alternately, we could follow one person. We could pick one. Dedicate ourselves to that person’s feed, and really pay attention. So if it were one, who would you want to follow?
LEVITT: Lindsay Lohan.
LEVITT: Who would you follow?
DUBNER: Who would I follow? Tell you what I would do. I’d auction it off. I’d say, we haven’t followed anybody. It’s time for us to follow someone. What’s the highest bidder?
LEVITT: The strangest thing to me about Twitter is I’d never been on Twitter and I went on and I had a Twitter account and it had one tweet. And it had my picture and it was from me and some person faked it. I don’t know why they stopped after one. But they only did one post. They still had you know a couple thousand followers from that one post, so, maybe it’s time for that person to get busy and start doing some more posts as me.
DUBNER: What was the tweet that the fake Steve Levitt tweeted?
LEVITT: It was the traditional first tweet. “Here I am. Time to get going on Twitter.” Something like that.
DUBNER: So, what does that say to you though? That you have a fake Steve Levitt out there who makes one totally worthless tweet and it gets a thousand or two followers, what does that say to you about the value of time people engage in the Twitter atmosphere?
LEVITT: Well, I’m offended that the guy didn’t do more posts. I want to see what I have to say.