The Social Media and Politics Podcast is a podcast bringing you innovative, first-hand insights into how social media is changing the political game. Subscribe for interviews and analysis with politicians, academics, and leading industry experts to get their take on how social media influences the ways we engage with politics and democracy.
Dr. Shelley Boulianne, Associate Professor in Sociology at MacEwan University, joins the show to share insights from her research on how social media is impacting citizens' engagement in civic and political life. Dr. Boulianne discusses the findings of her meta-analysis studies, comparing the results of existing research in order to better uncover how social media is affecting citizens engagement with politics. You can follow her on Twitter @DrBoulianne.
Check out Dr. Boulianne's full research paper that we discuss in the podcast: "Revolution in the making? Social media effects across the globe".
[00:00:00] Michael Bossetta: Welcome to Episode 26 of the Social Media Politics Podcast, bringing you expert insights into how social media is changing the political game. I'm your host, Michael Bossetta, political scientist at the University of Copenhagen. You can follow us on Twitter @SMandPPodcast, and also our Facebook Page: Social Media and Politics Podcast.
[00:00:48] Thank you so much for tuning in. I'm going to be interviewing Dr. Shelley Boulianne, who is an Associate Professor of Sociology at MacEwan University in Canada. And as we alluded to a little bit in our previous episode, looking at social media use in the British elections, as academics we tend to really focus in on one aspect of social media and politics. We'll look at one platform for one specific case or one specific protest. And what's cool about Dr. Boulianne's work is she's been looking at, in some cases hundreds of studies, and looking at the findings of those and then seeing what are the similarities and differences between those studies. And what really do we know about the effects of social media on for example: people's likelihood to vote, or people's likelihood to engage in activism or community involvement?
[00:02:40] My guest today is Dr. Shelley Boulianne, Associate Professor in Sociology at MacEwan University. Dr. Boulianne is an expert on how digital media influences civic engagement and political participation and joins us via Skype from Edmonton Canada. Dr. Boulianne, Thanks so much for taking the time out and welcome to The Social Media and Politics Podcast thank you very much for inviting me. So as I mentioned, in your research you're interested in the influence of digital media on civic and political engagement. Could you lay out these concepts for our listeners? What exactly is civic and political engagement?
[00:03:12] Dr. Shelley Boulianne: Sure. My working definition of civic and political engagement includes activities designed to influence government activities designed to improve community life and activities designed to express one's views about civic and political issues. So the type of activities that tend to link up with these studies are activities such as: voting, participating in boycotts, protesting in the streets volunteering in the community, talking politics, and then donating to political campaigns, charities, political causes.
[00:03:43] Michael Bossetta: And is it actions from only citizens or are there other actors that can engage in these type of activities?
[00:03:51] Dr. Shelley Boulianne: Well the research that I'm trying to summarize or is largely focused on surveys of citizens.
[00:03:57] Michael Bossetta: And speaking of that research you've done a number of really great what's called meta-analyses, looking at how social and digital media influence these type of engagement. So can you describe what a meta-analysis is and kind of how you go about conducting one?
[00:04:13] Dr. Shelley Boulianne: Sure. A meta-analysis is a systematic analysis of existing research on a topic. Specifically, I like to call it a quantitative analysis of quantitative studies on a topic. It's largely used in the medical and health sciences, and it's often used to summarize a large body of experiments on a particular topic. So I and others have adapted meta-analysis techniques to try to summarise survey research on specific topics. So like other meta-analyses I search for manuscripts through academic databases. I also use Google Scholar to find sources. And finally I consult the reference lists of relevant works to identify relevant research.
[00:04:53] Michael: Ah ha, so You look at a study and then who else has cited that study to sort of broaden that net.
[00:04:58] Dr. Shelley Boulianne: Sure, and then which sources have that study cited. So we're always building on each other's academic research and so I go forward and backwards with the citation lists. Who have they cited? And who is citing this reference, using Google Scholar and other similar software.
[00:05:15] Michael Bossetta: And then how do you go about it from there? So you have this this huge pool of studies. I mean do you go through and read each one individually or do you look at only there their findings and results?
[00:05:26] Dr. Shelley Boulianne: So my focus within that area of research is on survey research. I mean there is a lot of studies on social media engagement that are rich descriptions about specific election campaigns or protest events, and so that type of literature is really hard to summarize because it uses very different methodology. And I have it on my To-Do list to try to write some sort of comprehensive lit review of those types of research projects. But for a meta-analysis it really is meant to be a quantitative review. And so my research is largely focused on survey research. So with these latest studies that I've done have actually hired students to go through studies and figure out whether they use survey research and whether they test the relationship between social media use and engagement.
[00:06:15] Michael Bossetta: Interesting. So do they actually scout the research as well or is it you who do scouts the research and then passes it down to them and has that kind of filter out what's relevant and what's not?
[00:06:26] Dr. Shelley Boulianne: I would say it's a combination of things. I do like to send them to the academic databases to do searches based on keywords and to consult within the academic databases. And then in terms of Google Scholar we're all sort of hooked in to that, you know to find updates on things that we've queried about or we find updates on what's been cited in this area of research. And so we get those regular updates. So it's a lot of back and forth with my research assistants. Sometimes you find the same study on the same day and we're sending it to each other saying look there's a new study. So I guess a it's team effort in terms of finding those studies.
[00:07:05] Michael Bossetta: Right. And what's the... It's an interesting method for me because it's not that you're, you know, going out and conducting surveys yourself. Even though you do that and other research. But what's the benefit of this meta-analysis method, versus using survey data from a specific election or going out and conducting your own?
[00:07:25] Dr. Shelley Boulianne: Well, definitely. I mean, any single survey is going to have its limitations. We can only ask so many questions. And then of course our sample size will be very much limited by budget. And so I argue the benefit of a meta-analysis is that it basically takes all of these surveys, all of the people who have responded to these surveys, all those survey questions that have been asked within these surveys, and compiling all of that data together in one data set to try to summarize what is the big picture around the relationship between social media and engagement.
[00:07:57] Michael Bossetta: And so let's get into that in a little bit. In 2015, you published an analysis of 36 studies that looked at how social media influences political participation. And so my first question is why did you choose to focus on that topic specifically? What's interesting about the connection between social media and participation?
[00:08:20] Dr. Shelley Boulianne: Well, that paper came about in a rather odd way. I was invited to speak at the 2014 American Sociological Association Annual Meeting. I hadn't planned to do a meta-analysis, but one of the conference organizers read my 2009 piece around internet use and engagement, and she wanted me to give an update on the findings. And so what we had agreed upon and what we could do and the timelines that we had, I decided to focus very narrowly on social networking sites and what this new research on social networking sites was saying about engagement and that's how I got back into I guess the meta-analysis business. I was invited to update my findings, and I decided to do that very narrow focused.
[00:09:03] Michael Bossetta: And what was the timeframe of those studies that you were looking at?
[00:09:07] Dr. Shelley Boulianne: Those, because I focused on social networking sites, I would say I have just maybe one or two studies that was in 2007, possibly one in 2006. But the bulk of the research was collected in 2008 through the 2013 time period. And then of course the paper was published in 2015. So there's always a little bit of a time lag between when the paper the meta-analysis gets published and then when the data has been collected in these other sources.
[00:09:37] Michael Bossetta: Yeah it's interesting. 2008 is kind of an early adoption period. I think especially for scientists that are that are looking at this, so you have the nice, uh, catching the beginning there. So what did you find? Was there, you know, some common findings that came out by looking at all these 36 different studies? Or was it really a kind of grab bag of different findings?
[00:09:58] Dr. Shelley Boulianne: Well I mean, my intent with that piece was really to just give a quick update on what was happening with that research on social networking sites and engagement. So I had a rather narrow focus or a narrow intent with that research. Which was just to give an insight into what was happening. And so that piece really focused on what types of social media uses matter for what types of engagement. And also to address the broader question that was going around, which was: Is social media having a negative effect on engagement? Because that is basically what the concern was at that time when I was writing. And so my objective was to address that point. Is it having a negative effect? And I think pretty conclusively said that it's not. I mean, most of the studies we're coming up with positive effects are seeing a positive relationship between social networking site usage and engagement in civic and political life.
[00:10:53] Michael Bossetta: And what exactly does that mean? A "positive effect." We have some students listening to the show who may not be so caught up in their effects research. What is it that that increases that participation or has that positive effect?
[00:11:06] Dr. Shelley Boulianne: So I think that that piece really highlighted two different aspects of social media use that we're having positive effects. One is that social media was building people's social networks. So, you know, we go to social media or social networking sites, we're finding friends, we're building up those networks, and we know from research from the 1950s that building your networks increases the likelihood that you get asked to participate in civic and political life. It increases the chances that you will in fact participate in civic and political life. And so what I think that 2015 piece illustrated is that this is happening online now. It's happening through these social networking sites. People are joining groups, they're building their friendship networks, and these networks are becoming mobilized. So people are more likely to be engaged in civic and political life. And then the people that are engaged in civic and political life are also using social media to talk about their activities. And so we see that sort of dual process happening, based on that research.
[00:12:08] Michael Bossetta: And, was it so much that the networks corresponded to one's offline network, or? Because there's been some talk about social media enabling people who had known each other previously to kind of seek out and form networks around similar issues or what we're calling "issue publics." Is that the type of networks you're talking about? Or is it more about mobilizing your friends who you already know.
[00:12:33] Dr. Shelley Boulianne: Yeah I think that this is where the research needs to get into more of the nuances. What type of networks do you create in your social networking sites? And I think part of the question is whether or not your social networks are different across different platforms. So we know quite a bit about Facebook research. Because still, the large majority of this research is done focusing on Facebook. And so, the networks we create and maintain on Facebook are very much different than other platforms where we might interact with people that we don't know as well. We have weak ties too. So I'm glad to see that the research is actually moving in this direction of finding out what types of networks are we creating and maintaining in these social networking sites. Because I think that the different types of networks are going to have different outcomes in terms of whether or not we become more engaged in civic and political life. But again, my research is responding to what's out there. And I think that there needs to be more research on the nuances of what types of networks we're creating in these spaces.
[00:13:36] Michael Bossetta: Right. Because that might be a little bit difficult to get out with with survey data or the predominant survey approach needs to be updated. But I want to ask you about one of the difficulties that you noted in that paper, which was that the different research designs and methods can make it a bit difficult to compare the results of these studies. So can you go into that a little bit in terms of where the difficulties in comparing these studies it sometimes might be very different even though they're asking the same questions?
[00:14:04] Dr. Shelley Boulianne: Right. Well I would say that the majority of the studies use cross-sectional research designs, which is basically you interview a group of people and you look at what they said about media use, and what they said about engagement, and then try to make some sort of assessment of: is there a relationship? And so, for those studies it is much easier and straightforward to try to summarize the results and to say look these two things, these responses to the media use questions and responses to the engagement question, are sort of going in the same direction. And the direction usually is, the more you use social media, the more that your reporting higher are levels of engagement in civic and political life. So these are rather straightforward to deal with them the analysis techniques are pretty standard. And so this was quite easy for me to do in terms of summarizing. What gets complicated is the longitudinal designs. And I would argue the value of the longitudinal design, despite the complexity that it introduces, is that there is great value in the longitudinal design. So the designs that we're seeing, they track people over time to see if there's changes in their levels of engagement, and whether these changes can be related to changes in social media use. And so it's a much stronger set of data to try to think that there's a causal relationship - that on is causing the other. But the problem is there's a much more complexity in these types of designs. They tend to be limited in terms of their sample size. They tend to use student samples because they're the easiest ones to track over a period of time. And so these complexities of the sampling approach, the complexities in the analysis approach, that makes them very hard to summarize the result.
[00:15:46] Michael Bossetta: Yeah, and I would think that it's a little bit difficult today, where you have, as you're saying there's a lot of focus on college students, using Facebook when Facebook first came out. But now what we see is some, you know, 15 and unders don't have Facebook. They might be just on Instagram, and then older people are starting to come on Facebook, and there's tons of new platforms to cope on. So I think it's, it will be difficult in the future to kind of disentangle that over time because of all these external factors that are in the social media landscape.
[00:16:21] Dr. Shelley Boulianne: Right. And just to add to that the more platforms, I mean, this will be great. I can respond to the differences by platform once there's research out there, when there are multiple studies out there looking at different platforms. But I guess the problem is that, if they are using multiple platforms, how do we figure out: So this is you know an Instagram effect, and this is Facebook effect, and this is a Twitter effect? That makes it very hard because a lot of these people are using multiple platforms.
[00:16:48] Michael Bossetta: And do you think there's, by looking at self reported survey data, I mean do you think there's a need for more studies looking at the actual metrics of social media? Or, is there a way that you can incorporate that, or complement that, by survey data? Because you know, one of the problems with surveys is that people say they do something, and maybe it's inaccurate what they're actually doing. So what are your thoughts on that?
[00:17:10] Dr. Shelley Boulianne: Well I absolutely love the idea of mixed methods research. My own research you mentioned, I do surveys. Well, I do surveys, and then I supplement the survey data with other sources. So for example, the projects that I'm working on right now looks at social media use. So many of these same survey questions that I've seen used in other studies, I look at social media use measures. I look at engagement measures, and I'm looking at it around a particular event, which was charitable donations in relation to a wildfire that my province had last year. And what we're doing is we're using the survey data, and we can establish that there's a connection there. That there's a correlation, or a relationship. The more you use social media to find out news about this event, the more likely you were to donate to the Red Cross and other similar organizations. But what we did, that I think should be done more often, is we looked at what was happening on Twitter. What was the Twitter discourse around that time period? And what you're seeing to supplement what the survey data was saying. What you're seeing is these messages of "help out," "donate." You know, "these people are suffering." You know, "we need to do something." Those sort of calls to action were evident in the Twitter data. So we use the combination of sources, and I think that that's really where research should be done. I know that this is done often in book format, where you see survey data supplemented with social media analytics, but I'd like to see more done in the article length because I really do think that surveys are only so good for making that theoretical connection between two variables that there really needs to be details about what is the content that is circulating.
[00:18:53] Michael Bossetta: Right. So it's not necessarily so much that people are just using social media and then happen to organically donate to this cause, but that there are actually certain things going on on the platform, like asking people to donate, that would then elicit more of that response.
[00:19:09] Dr. Shelley Boulianne: Exactly.
[00:19:10] Michael Bossetta: OK. So, let's fast forward to today maybe focusing on the 36 studies that analysis is not doing proper justice to a more recent work where you've looked at 150 different studies on social media and participation. And what what changes did you observe if any between looking at 36 studies in a more updated list of 150?
[00:19:31] Dr. Shelley Boulianne: Right. So, one of the things that I saw right away when I started looking at this literature is that what people are focused on in terms of social media use has changed. There is a lot more studies that look at the use of social media for political expression or for discussing political issues. And so there was enough of those studies that have focused on this that I can really look at that particular finding, and look at whether it had outcomes on people's offline engagement in civic and political life. So that was one thing that was new with this area. Again, these meta-analysis have to evolve based on what the research is producing, and the research was producing some consistent findings around this idea of political expression on social media and offline forms of engagement.
[00:20:20] Michael Bossetta: Ah ha. And I've been looking at a few of these studies where they'll say, for example, that those who are more likely to tweet during a political event are more likely to go out and vote, or go out and canvass, or campaign. And I'm wondering, I mean, does that have anything to do with that the people being surveyed or already politically interstate? Or, how do you disentangle that from the causal mechanism of social media to go out and do something? I mean is that something that's clear from from this 150 study meta-analysis?
[00:20:56] Dr. Shelley Boulianne: I think, again, this is where the value of a longitudinal design comes in, because we can account for things like are they already politically interested. We can ask that question and try to account for, you know, is it political interest that it's causing both does observing these political events through a second screening and whatnot, and also causing engagement in political life? I mean, we can try to do that, and try to untangle that process using the cross-sectional data, but really the value is in longitudinal studies or experimental studies. If there were studies looking at changes in behavior, then we could untangle the causal process. I don't think that we're there yet. I mean the longitudinal studies that we have really aren't able to untangle that process just quite yet in terms of what is causing what.
[00:21:49] Michael Bossetta: And are there any other differences that you observe between... So in the 2015 study you found this positive effect on social media and participation. And then, in the updated version, is it similar positive effect?. You know, I'm thinking about these echo chambers and this kind of reinforcement idea. Did you find evidence for that or was it still a generally positive trend?
[00:22:14] Dr. Shelley Boulianne: So I definitely still find that positive trend. I guess the reason that I set out to do the second piece is that one of the questions that I couldn't properly address were that 2015 paper is the magnitude of the relationship. And so I look at, yes, there's positive effects, but only half of them were statistically significant. And I know researchers are always concerned about, is it significant? So the question I ask with this new study is: Are the effect sizes substantive? Is it really a game changer that social media use is going to cause changes in things like voter turnout? Is social media use going to change someone's decision about whether to participate in a street protest, or whether to volunteer in their community? And so I needed to look at the sizes, to look at that magnitude of the relationship. So that was really the primary objective in starting out the second piece on this meta-analysis of social media trying to get at: what is the magnitude of the relationship? And in particular, the discourse isn't debating anymore whether there's a positive effect, it's debating whether or not there's a substantial effect. And so that there's a debate about whether the effects are revolutionary. And so that's the title that I have for this paper, is trying to assess whether or not the effects at least when we're talking about citizens engagement in civic and political life, are these are facts revolutionary? Are they game changers?
[00:23:38] Michael Bossetta: And, have you gotten to that final verdict at this point or are you still working through the data?
[00:23:44] Dr. Shelley Boulianne: I would say that there's certain aspects where we can see that there's revolutionary effects. And the aspect that I think is most evident is around the political expression on social media. We see it in the survey data and I think that the analytics of social media data is showing that same pattern. And so I see that there's something possibly revolutionary in there. And the other dimension, or the other really that I'm analysing this data, is to try to understand how the effects differ across different political contexts. Because I think that's part of the answer of whether these effects are revolutionary, depends on where we're talking about these effects occurring. So social media affects in a media rich environment, or in a system where there's a free press, I mean social media effects are going to compete with other media effects. Whereas we see in the meta data, when you look at systems where there's a lack of a free press, we see a much larger relationship. And again it's social media that's filling in a huge gap in terms of information needs, and it's causing a larger effect in terms of peoples' engagement in civic and political life.
[00:24:53] Michael Bossetta: So, that might be something like a country where they don't have a free press, maybe an authoritarian regime, but social media sort of has this potential to create awareness about something that may spill over into a protest that's organized on social media.
[00:25:09] Dr. Shelley Boulianne: Yes, absolutely. And I would say that it's evident in systems where there's not a free press or the authoritarian regimes. But also there's a strong relationship in terms of systems where there's a transition in democracy or there's a partly free press. So, you know, it's developing a democratic system but it's still in those developmental stages. And you see strong effects there as well.
[00:25:31] Michael Bossetta: So kind of dovetailing on that, and it's kind of a loaded question, but based on having looked at all of this data, what's your assessment in terms of the potential for these social media in terms of impacting the quality democracy? Are they positive or negative for democracy? Or is it, you know, the argument that they're neutral and it's how you use them?
[00:25:53] Dr. Shelley Boulianne: Well, I guess my first reaction to that question is that we need to look beyond democracy and the role of social media beyond democratic systems, because as I mentioned the effects are quite substantial when we're looking at other types of political systems. And so asking about whether it's good for democracy, you know, that's sort of a different answer because it is good for citizen participation in all types of political systems. And the question of: Is it good for the quality of democracy? I guess that sort of depends on what you think is good for democracy, and I think more citizen participation is a good thing for democracy. So in that line, or on that note I would say that social media is having a positive influence on democracy as well as non- democratic states.
[00:26:40] Michael Bossetta: Good answer. And just wrapping up, can you give us a teaser about where your research is heading at the moment? So you mentioned a study about wildfires and the effect of social media participation on that. Anything else we should look out for?
[00:26:56] Dr. Shelley Boulianne: Yes, along the lines of the meta-analysis. I have decided to revisit the 2009 study that I had published around internet use and engagement. And so I have a database there of over 300 studies...
[00:27:09] Michael Bossetta: Jeez...
[00:27:09] Dr. Shelley Boulianne: Yes, I'm still collecting the studies. For that one in particular I want to answer the question what this new research about whether the 2016 period is distinctive. Whether the U.S. election in 2016 will produce different outcomes in terms of citizen engagement. But I'm also looking more around the world and there's a lot of Western democracies having elections in 2017, and so I have this database, and I have some ideas about what's happening in the database. But I'm really looking at these these brand new studies to add to the question of are we seeing something dramatically shift in terms of media and its role in citizens engagement. So that's one of the objectives that I have over the next few months. The other one is to look at the effects and how they're distinctive for youth. Because I think this has been a reoccurring theme with my meta-analyses to say that youth are distinctive, the effects are possibly larger for youth. And so I've decided to tackle that question: how does social media and digital media, how does it have a different affect for youth compared to other age groups?
[00:28:17] Michael Bossetta: Very interesting, very relevant, and a huge knot to untangle though. So, best of luck with that, I'll be looking very much forward to reading the final results. And Dr. Boulianne thanks so much for coming on the show. Appreciate your time.
[00:28:31] Dr. Shelley Boulianne: All right, thank you very much Michael.
[00:28:33] Michael Bossetta: I've just been speaking with Dr. Shelley Boulianne, Associate Professor of Sociology at MacEwan University. You can follow her on Twitter @drboulianne.
[00:28:46] All right, that's a wrap for this episode of The Social Media and Politics podcast. Hope you enjoyed the show, and thanks so much for tuning in. Next week we'll be speaking with Sam Jeffers co-founder of the group Who Targets Me, and a former executive director at Blue State Digital.
[00:29:02] Feel free to connect with us on Twitter @SMandPPodcast podcast. Direct any questions, feedback, suggestions for future episodes that way. If you want to be a hero leave a review on Apple Podcasts and help us game those algorithms. We're climbing thanks to you guys. Keep downloading. Keep sharing. Keep learning. I'm your host, Michael Bossetta, signing off from off in Copenhagen. See you next time.