Using Big Data to Solve Big Questions in Sociology | Interview with Dr. Omar Lizardo

0:00:12.6Big data

Jed Macosko: Hi, this is Dr. Jed Macosko at Wake Forest University and Academic Influence. And today, we have, visiting us from UCLA, Professor Omar Lizardo, who is a sociologist. So a lot of things must have changed during your career as a sociologist. The thing that I’m most curious about is big data. We hear about big data a lot and you have done a lot of studies using old-fashioned data, survey data…

…how has the new type of data changed things, and from a standpoint of a sociologist, has it even changed that much? So tell us about that.

Omar Lizardo: Yeah. No, yeah, that’s definitely a great question and it’s definitely one of the biggest changes in the field, just in the last 10 to 15 years. So it’s been interesting to be in the field as those things are changing and I think that the... My analogy is with the old-fashioned data, the survey data, I think that they were always thought of as like you looking at society from the top-down, like you take a probability sample and you’re trying to capture social trends, and it’s got that aspect of it, but it’s always, but it always... And I think with survey data, it always feels like you’re [inaudible] add and remove. You get very gross distributions or percentages about [inaudible], etcetera.

And I think that switching or going to this more big data, even though it’s called big data, it’s actually, I see it as more like finer grain data. And that’s why it’s big; it’s not necessarily it’s big, it’s big because it’s got a lot of information, but the information is actually a much finer grain than you would get from a survey. You can get a whole, somebody texted at what second of the day. It makes it big because there’s a lot of seconds in the day. [chuckle] So I feel like it’s, the analogy is like looking at something from the very top and getting very broad, that’s the old-fashioned data.

The thing about the big data, it allows you to kinda zoom in, it’s almost like a microscope, at finer grain levels of human behavior. So you go from an attitude or a self-report, what you did last week or last year, to like, "This person texted this other person at this time of the day or night," or whatever. Or you... If you go for the big data, that behavioral data, that actually, quickly sometimes, can track where people are at, you can track human mobility. So I think that that’s, in many ways, that’s a finer grain description of human behavior, like, "I went from this room to this other room," than you would get from the old.

So I think that the old versus new, I think the analogy that I use is more like a zoom in versus zoom out. And sometimes both perspectives are important, and I think that... And of course, in the social-behavioral sciences, it’s really also a move from relying on self-report data, essentially people’s perception of what they do and what they want to reveal about what they do, to this more behavioral third-person data, which was just... A lot of the so-called big data in the social sciences is really data that monitors human behavior at this very fine scale, both temporally and in terms of the scale at which things get collected. So my sense is that I feel like it’s when I do research using old-fashioned data versus data like that I feel like I’m moving back and forth in terms of that, like zoom in, zoom out on people more.

Jed: So, you do use big data in your own research, is that what you’re saying? And if you wouldn’t mind, tell us a little bit more about that.

Omar: Yeah, so I’ve been at UCLA for the last three years, I’m just relatively new. But before that, I was at the University of Notre Dame for 12 years. And there, University of Notre Dame is an interesting place because it’s a research university, but not as gigantic a separation as something like UCLA. So, in many ways, it’s kinda like a small campus, but there’s a strong culture of research and in that respect, there’s also less strong barriers between departments, so and there’s a lot of interesting interdisciplinary centers. One of the ones that existed there when I first came there in 2006 was an interdisciplinary network center that was founded by now a very famous network scientist named Albert-László Barabási , well known for lots of other things.

He is now elsewhere, but I think that the interesting thing there is because it was a network center, it didn’t have the disciplinary tag, so anybody who was interested in networks, from biology or sociology, etcetera, could come in. And there I started collaborating with some people that were interested in this newfangled social and behavioral data as connected to network analysis. And that’s why I started doing some projects, using what now gets called big data.

The earliest one had this crazy... Well, then at the time, now it’s not unusual, but at the time, I thought it was crazy because it was essentially the entire data dump from a Spanish cell phone company that had essentially data from six months of all their customers and it ended up being 28 million people. And of course, if you... 28 million people and the number of edges between [chuckle] those people, and so it became like some gigantic order of magnitude higher. So incredible amounts of information at a very... It wasn’t in a very fine grain scale. And we were like, "What do you do with all of that?" [chuckle] That’s a... Which ended up being really interesting to work with physicists and other people that were also interested in the same, similar questions.

Jed: So you were working alongside physicists with the same data dump from the Spanish telecommunication company?

Omar: Yeah. The first collaboration had one physicist, Zoltán Toroczkai , who’s the Notre Dame computer scientist, of course, and interested in data mining. Nitesh Chawla is also under him, so... And then there were a couple sociologists, myself and another colleague. We applied for a grant to analyze this data from this more social science oriented perspective. And to me, it was essentially like going to grad school again because collaborating in that sort of interdisciplinary environment, you learned new... You learn newer ways of looking at data, but there were new ways...

You learn how to look at data in a different way than you are trained in [inaudible] which is based on the idea of like, you’re gonna be sampling from a large population, you’re gonna be looking at small things, and so a lot of [inaudible] inference, versus here, it’s like, well we just computed a mean and it’s 20 million cases, so that’s the number, there’s no... So you get out of that mode of thinking about uncertainty and much more thinking about the interpretation of very basic quantities. And so in many ways, it’s a very... You find it simpler because you’re not looking at all the statistical things, but it’s also more challenging because then you’re forced to really think about what the underlying mechanisms that generate even a very simple observation might be.

And the other thing is, in the social sciences, obviously, your background in physics, you know this, it’s just a much more visual culture of data analysis, and that was new to me. Essentially, you are looking at a plain graph, a picture of a distribution, that’s really what you’re trying to explain, not just some coefficient in a regression model. So to me, I think I’ve learned a lot for that early introduction to that kind of interdisciplinary culture of data analysis, which has been really important for me going forward.

0:07:51.8Urban to rural

Jed: Well, I started this conversation just ’cause I wanted to know in general how big data has influenced your field, but now we’re going into this very specific case, where you had this data from the Spanish telecommunications company. I wanna know…

…what did you guys find out and was it interesting from the sociology point of view, from the physics point of view? I mean, what are some really interesting take home things that our audience might find interesting?

Omar: Sure. I mean, there was one project where we were the... One of the sociology grad students were able to geocode the data into some regions in Spain. And one of the questions that we were able to ask was, it’s an old sociology question, this idea that whether there’s a correlation between urbanism, the center, which places... Has high population density and urban concentrations, and the quality of social ties, usually defined as this idea that you have strong ties versus the ties weaken with urbanization.

And what we found was essentially that, at least the old sociological theory that postulates a one-to-one correlation between urbanism and the decline of strength of ties, at least as measured by frequency of activation, in terms of phone calls and text messages, this wasn’t true for this data. That [inaudible] it was essentially either no... A very weak effect of urbanism, or actually, the reverse in that... In that in a lot of urban areas, people actually had a lot of strong connections with people inside of that area. And in rural areas, you got weakening of internal ties, but strong ties to people outside of that area. It’s an interesting... Using that newfangled data to answer a really old sociological question that people have been asking since like the late 19th century. That was interesting.

Jed: So since the mid-19th century, people have been thinking that when you move into the city, you lose some of your close ties.

Omar: Correct.

Jed: That’s what the theory was and the data from Spain showed that that’s not true. At least if you only look at their texts and phone calls. I mean that’s all you had. You didn’t have any data about face-to-face interactions.

Omar: Right. So we were using essentially just frequency of activation and inter-temporal differences in how, thinking of activating a social tie as a proxy for strength. Yeah, that’s right.

Jed: How do you correct for the possibility that, well maybe the people that are in the rural areas are just meeting more face-to-face and that, that does make it a significantly more close connected society in the rural areas once you factor those face-to-face ties in? How did you, could account for that?

Omar: Right. So that’s something... So because we didn’t have access to the more face-to-face interactions, we were presuming at least, that the regular face-to-face interactions wasn’t necessarily gigantically different between rural and urban areas, so that at least with the more activation of ties via telecommunications, at least that difference, right, would... If there was a difference between rural and urban areas in that respect, at least it would be... [inaudible] at least, some kinds that there’s not a decline necessarily of those core ties in an urban area. Presuming, of course, that those face-to-face interactions do remain in both places, inclusive of the rural areas.

Jed: Were you able to cite some other studies that show, yeah, there’s not much difference in face-to-face interactions with close friends? ’Cause I mean one might naively assume that when you move into the city, the people that you interact face-to-face with on a daily basis are people that are not really your friends, and so there might be a bit of a loss of that close, close interaction, because in the rural area, if where you’re from, you might have seen all your friends every day on a face-to-face basis, but it gets diluted when you go into the city. Is there any study that showed that you could kinda say, "Well, they’re about equal"?

Omar: Well, yeah, so what has been shown is that, that parallel you’re describing is more or less right, in the sense that the ties that you see frequently in urban areas are not necessarily your close ties, but what has also made it shown is that what happens in urban areas is that essentially, you get a dislocation between strength and where your ties are located, so that you can have very strong ties, even though geographically, they are farther away than they would be in a rural area.

So this idea that essentially, one of the sociologists at the University of Toronto named Barry Wellman created this evocative imagery. He called, your community has become personal, and essentially instead of being be tied to place, they become tied to your ego network essentially. So you have a personal community, strong ties, that you then, you rely more on this telecommunications technology to keep alive. And you have, in many ways, you become more reliant on the technology. And I think that, that’s definitely part of the pattern that we’re picking up with the telco data, in that we did find urban areas had more strong ties within the urban area, which is a version of maybe of this idea of dislocation of ties from place versus in rural areas, there might have been more of a correlation between time and place that maybe also makes them less reliant on telecommunications technology to keep alive.

Jed: So your study leaves the possibility open that the theory developed in the mid-1800s was true, it still leaves that possibility, that there are better more ties in the rural areas, but it would have to be because of a huge or a significant amount more face-to-face interaction.

Is that true? Or have you pretty much ruled out, at least for this data, at least for Spain, that the theories of the 1800s doesn’t hold true?

Omar: Oh, yeah, I think that that half, you can probably make the case, if you were to equate high quality ties with close, in the very traditional sense, close, face-to-face, neighborly, etcetera, that maybe you do have a difference in that quality when you move to an urban area. But I think that if you were to... The more extreme version of the theory, which is all ties become weak and decline, etcetera, then that we can rule out because essentially, what happens, we begin to piggyback on the telecommunications technology to actually keep some strong ties going.

Jed: Good, so you were able to at least revise that formula from the 1800s that said, "Every single one of your ties gets weaker." You were able to say, "No, that’s not true."

Omar: Right.

0:15:03.8COVID

Jed: Okay, well, that’s really interesting. Now, this of course brings us right to COVID because we’re all dealing with a situation where we can see people less face-to-face.

Have you or your colleagues studied the sociological impacts of COVID? Is it too early to do that research? Or what do you think?

Omar: Yeah, no, that’s a great question, I think it’s early. It would’ve been nice if we had one of those ongoing data collection projects and I’m sure some people probably got lucky and were able to do that. We didn’t do that. But yeah, but I think that is an important question because I think it’s similar to this idea of, what happens when there’s either exogenous or organically generated shock or constraint on your capacity to do face-to-face interaction, which is very closely connected to this classic contrast between urban and non-urban settings, is that urban settings themselves are, put impediments on face-to-face interaction.

Even before COVID, remember, I moved from South Bend, Indiana to Los Angeles, and one of the things that my wife always said, it was just so much easier to get together with people in South Bend, because there was driving, you can just think [inaudible] we’re literally closer to people. So even before COVID, in an urban area like South Bend... Like Los Angeles, puts restrictions on face-to-face interaction, and exogenously generated by the fact that... Large space, driving, etcetera. And I think that now with exogenous shock, a completely exogenous shock is that times a thousand, which means that then, then I think then this substitute, which is this Zoom or other telecommunication then become essentially the lifeline that you have to keep your ties going.

And in many ways then, that will... I think that that will result in even more inequality in the capacity that some people will have to sustain ties because if you make something reliant on technology or you’re making something reliant on the cultural capital that is required to even be familiar with the technology, and so... And then I think that, my guess would be that there has been... Everybody has experienced a shock, so you’d get a general decline, but then you also get kind of a slope connected to things like age, class, income, etcetera, because all those things do help you, at least, do the substitution that’s required to keep those ties alive.

Jed: So what you’re saying is that people in different age groups, different socioeconomic groups will experience a greater decline in their social connections, and it really comes down to how comfortable they are and how much access they have to the appropriate technology, like the Zoom call we’re doing right now, is that what you’re saying?

Omar: Exactly. Yeah, we knew from even, [inaudible], before COVID, that there’s strong age, class, even racial and ethnic gaps, and those basic, web and technology savvy, that do account sometimes for even gaps in capacity to use the internet to do basic things. And I think if you increase, if you then make social interaction dependent on those things, then you’re essentially piggybacking, you’re taking that inequality that was purely based on instrumental use of technology and now it becomes inequality based on human contact, the capacity to interact with people, and that definitely exacerbates things. So I wouldn’t be surprised if a lot of those inequalities then get translated in terms of inequality of connectivity. And we know that even those group... The same groups that I mentioned, are also the same groups, especially the elderly and other people are also the groups that are also more... Most at risk of being socially isolated and lonely. You get an exacerbation of those dynamics, I think, through COVID.

0:19:06.5Helping the world

Jed: Does your own research help, in some ways, to undo those inequalities? You’ve gone into sociology because you found it interesting, but does it also give you an avenue to bring about some kind of positive change to the world around you? And maybe we can just end the interview on some of the things that you see, your research as... And don’t be afraid to toot your own horn here, you are one of the most influential sociologists in the world right now. So please tell us how does your field, ’cause it bears on all sociologists, not just your own research, but tell us how does your research help the world?

Omar: Well, not a question that I usually ask, but I think that... I think that’d be, I study connectivity between people and I think that connectivity is one of those things that has essentially been connected to pretty much every outcome that you can think of: Health, wellbeing, longevity, etcetera.

I also study cultural consumption, which I think it’s one of those things where people usually didn’t think of that as a good or necessarily as a general good, but I think it is, mostly because I think that connecting to culture allows you to connect to people via culture, and I think that there’s, in many ways, there’s an intimate connection between those two things.

In fact, there’s a... But I also study inequality between both those, the inequality in terms of social connectivity, which is a long-standing thing and inequality in terms of cultural access and cultural consumption. And I think that... One of the things that I found repeatedly in my work is that cultural access leads to social access and there’s even a nice mutual loop between those two things.

And I think that one of... The projects that I’m... It’s an early... It’s my first early COVID-related project is about what has happened to people’s cultural access in the wake of COVID. And I think that... One of the ways in which some of the things that I found can be helped to change the world or to help the world is by pointing out the importance of this idea of how do you make culture accessible to people? Because I think that sometimes gets overshadowed by a lot of other inequalities. And usually, in terms of more material resources that people think about, but I think that cultural access is as important as some of those other more tangible resource inequalities. And I’m also thinking precisely because a lot of that cultural access then becomes tied to your access to what sociologists usually refer to as social capital, benefits embedded in social relations and connectivity.

So I think it’s one of... Thinking about that, and of course, the social inequality aspect usually gets talked about, but I think that thinking about the connections between that and cultural access becomes one of the ways in which you can make the world better by increasing general access to culture to people. And I think that, that’s one of those small lessons that I think from my research you can draw.

Jed: So you’re saying that basically, we just need to give everybody access to Netflix and then they could all be culturally accessing the world... I’m just kidding. [chuckle] But, no, seriously…

…what would you say would be recommendations about increasing cultural access? ’Cause I don’t even have... As a physicist, I don’t even have a good idea of, where is cultural access happening? If it’s not Netflix, where is it? Tell us.

Omar: That’s a great question. So Netflix is definitely one of those things, but it’s also a restricted thing. There’s a lot of, for instance... Because Netflix is one slice of where cultural production is happening, but if you think about other institutions that are being affected by COVID, maybe with fears or... That’s a lot of cultural access that it’s... Access to the visual arts, in addition to concerts, but also broadly defined to include not just popular culture, but the old symphony orchestras and things that... It’s an entire cultural field that is produced, not just in the for-profit arena, but also in the non-profit, and neighborhood and civic associations that produce culture. And one thing that I think that, that is, there is an aspect of isolation, that it’s a version of suffering and the [inaudible] provision that people are undergoing that... One thing, it doesn’t get as much of the press, but I think it’s ultimately important because it has a strong... It has a strong connections to both social attachments, but also other kinds of attachments.

0:24:26.4Social capital?

Jed: So for somebody like me who’s not in social science, I tend to think of these cultural things you just mentioned as things that could improve society by inspiring some young person as they go to a symphony concert or go to an art show or something like that. And that is what helps lift society. But everything you’ve been saying is that as you consume cultural resources like these, theater and things like that, it actually gives you social capital. And you were talking earlier about how, if you talk to your caddy because they’re outside of your social circle, you talk more about these sort of pop cultural things…

…are you saying that if you give access to these social consumables to people who are experiencing them right now at an unequal rate, they’re not getting them... If you open it up, then they can basically have more social capital because they can talk to their friends or talk to people outside of their community about more things? Is that how it works? Is that the mechanism by which they get more social capital?

Omar: That’s one mechanism. Essentially, the capacity to cut across social circles and social worlds, which is a version of that. There’s also a version that comes from simply just maintaining the connections that you have, essentially deepening the connections that exist in your world. There’s an interesting idea that I’ve recently been fascinated by, in social psychology, this idea of that one of the core human motivations is this idea of shared reality, that when we connect with other people that like to think that we share the same world and share the same beliefs, and whatever, and I think that one of the core ways in which we maintain and prune our social ties, is through a shared reality mechanism. And one of the primary ways in which we share realities is when we share this aesthetic or fictional realities, that I think that then become a strong signature for having a strong tie with somebody else.

And in many ways I think that, of course, the dark side of that is that when you have strong cultural divisions, people literally consume different cultural goods, then you get non-shared realities, right? But I think that the positive side of that is that when you have this more democratically shared access to cultural goods, then you’re able to break those and allow people to connect outside of their traditional social circles, which can become a good for themselves, but also just in general, social good too.

Jed: So if you were the Mayor of Los Angeles, what would you do to help open up these cultural resources to everyone so that there could be these shared realities, so that people who don’t have social capital could get more social capital, and lift them and their community out of some of the financial, health, economic problems that their community faces? What would you do?

Omar: I usually don’t fantasize about having political power, but I think that in many ways... Accessibility goes beyond economic accessibility. Right? I think obviously, that’s an important one. So I like the fact that a lot of the museums and other things in LA are free, but I think that you also have to make them accessible in a way that makes them less class-marked in the sense of seemingly, there has to be an accessibility barrier. I think connecting to people’s already folk cultures and tastes, for instance, to make those connections between the music that gets played at a symphony orchestra beyond the traditional canon, to include other forms of cultural expression become important because that makes that art form more accessible. And the same thing with visual arts and other forms of culture, I think that that also becomes more accessible if you connect with the already existing culture that people have.

And I think that... I would say that, in terms... Especially a place like Los Angeles, that has such a already vibrant multicultural tapestry of people. I think that going beyond the traditional canon and that sometimes gets this cultural benefit [inaudible] gets stuck with, will be an important way of increasing accessibility, signaling that you’re accessible.

0:29:19.2Sign-off

Jed: Yeah, well, this is fascinating. I can see, there’d be many more things you could do as Mayor of Los Angeles. But I can see which way you’re going with that, and I think it’s wonderful. And I really think it’s fascinating, the perspective and the helpful tools that a sociologist can provide to society to help our society become a more fair, equal place. So thank you for sharing all of this today, it was absolutely fascinating. I really appreciate you taking the time, and I look forward to hearing all the other great things that you do with your career.

Omar: Sure. Yeah, thanks, thanks for having me.