DW: Looking at political movements that have been associated with social media, whether it's the Arab Spring or Spain's 15-M movement, or more recently Ferguson and the #ICantBreathe campaign, can we really say with confidence that social media shape that kind of collective action? Or is it simply the platform on which these actions are played out - as well as on the streets?

Professor Helen Margetts: Social media does play a vital role. It makes possible something that is qualitatively new. It's allowing very tiny acts of political participation - such as liking or sharing, signing something digitally, or sending an email. All these are very small acts of participation which weren't possible before, because the transaction costs relative to the participation costs would have been too great. They're like tiny micro-donations of time and effort. They can be of money too, of course,... and they can scale up to something really important. So if you've got a million people on the street, we would argue social media played a crucial role in getting them there.

But if, as you say, social media lower the cost of engagement, then perhaps it's not worth as much? Liking something is easy, it takes even less effort than sharing something with your community. So are we really seeing a greater kind of engagement?

Obviously there is a variation in acts. So if you share a picture of a drowned Syrian boy, or if you share a picture of a sign saying "refugees welcome," the costs of that are higher than just clicking Like, because you're sharing it into your friends' and acquaintances' feeds and some of them might be upset by it or disagree with it, and they might unfriend you, so the costs are slightly higher.

Chain reactions

But I think the point is that this is part of a long chain of reactions between links.

If you think of the March of Millions in Egypt - leading up to that was a [Facebook] page called "We are all Khaled Said," which had reached half-a-million Likes by the time the Internet was switched off. Each of those Likes is sending a signal to someone else that this thing is a bit more popular, a bit more viable, and making it more likely that they will click Like and ultimately join the demonstration on the street. They're all part of a series of chain reactions. And what we're arguing is that they matter, because of this effect they're having on other people.

A lot of people say social media campaigns lack an identifiable leadership. But you do still need people at the front - be it the early adopters. You refer in the book to people with a "high locus of control," people who believe they control their own fate. Or people who through their support say to others that it's okay to be part of this. And in a way that does create a leadership.

But that's not what the data tells us. We expected to find that with the successful mobilizations, hashtags or petitions… If Justin Bieber signs up for something, obviously that's influential. But when we analyzed the data we found that it's still not as influential as huge numbers of people who don't have very many followers. It's them that are playing the crucial role, en masse. And that's what other researchers have found, too. So [a singer] might be involved, but I'd be very surprised if, when you looked at the data, he could be described in any way to be leading it. There will have been huge numbers of what some people call "critical peripherals."

Sustainable campaigns

What about sustainability of social media campaigns, because it's one thing to be successful, but a campaign has to have a sustainable outcome too.

Well that's the downside of what we're talking about. We're saying that most things fail, but the ones that succeed can succeed without the normal organizational trappings of a movement or a revolution, or leadership. That's why so many of the revolutions of the Arab Spring were disappointing, because they didn't have leaders or nascent political parties or organizations waiting in the wings. That's why the Muslim Brotherhood was so successful in Egypt, because it was the one organization on the civil society scene.

You do see it happen sometimes, though. In Spain you have seen what was the kind of movement we're talking about, starting to acquire organizational mechanisms in Podemos, which came out of 15-M. Podemos just existed on social media to start with. And now it's a major force in Spanish politics. It doesn't look like it's going away anytime soon.

Deluded by data?

There's been a debate for some years now about social media and its merits as a driver of democracy - a debate started I think by Evgeny Morozov [author of "The Net Delusion"], whom you mention in the book. Where do fit in in that discussion?

There was in some areas a hope that social media would reinvent the "Habermasian" public sphere and lead to a form of deliberative democracy that we've never seen before. And that was never very plausible to me. I don't think social media or the Internet are very good for deliberation. I think it's about the numbers. What we've got is these tiny acts that represent potentially important sources of data about democracy - what's working, what isn't working, what people want, prefer, need, and how they behave. There's lots of information in this newly data rich world which could make our democracy better if we can work out how to harness this new willingness to participate, rather than say, "Oh, God, it's just clicking buttons."

Well you've said before that we're "dripping with data," which makes me feel a bit odd, but it seems about right…

I think it is, because that's exactly what you're doing: you're dripping with data. That doesn't mean anyone's picking it up and using it, in the same way we don't use the water from a dripping tap. It's inefficient. But also we shouldn't feel it's a kind of conspiracy theory to get all our data. We're dripping with it.

Professor Helen Margetts is the director of the Oxford Internet Institute, which is part of Oxford University. She is also professor of society and the Internet, and a co-author of "Political Turbulence - How Social Media Shape Collective Action" (Princeton University Press, 2016). Peter John, Scott Hale and Taha Yasseri were her co-authors.