Monday, April 25, 2016

Who's Watching Webseries: A Case Study

Photo credit: Personal Space

I thought it would be cool to know how the audience for webseries differs from the audience for traditional television. There are a lot of things that prevent me from doing a real statistical study on that, so in the mean time, I looked at an interesting case study involving Battlestar Galactica and an upcoming webseries called Personal Space.

One of the most fascinating things about the emergence of webseries as a medium is the range of stories they allow space for. Webseries don’t have to get a greenlight from a network or studio, so webseries creators can be just about anyone—from seasoned professionals to people who just picked up a camera for the first time yesterday. And while that openness creates a vast range of production values and quality, it also creates a vast range of perspectives.

In particular, webseries seem to be more diverse than traditional television. I don’t have hard numbers on that—and the nature of webseries distribution would make unbiased data difficult to gather—but it’s anecdotally true from my experience. Whether it’s literary-inspired webseries creators swapping the gender, race, and sexuality of classic characters, or original webseries creators telling stories about corners of the world that don’t show up much on television, webseries seriously trump television, in terms of representation.

What would be interesting to know is whether diversity of stories translates into diversity of audience. And it would be really interesting to know whether webseries have more diverse audiences even if you control for the diversity of their casts, because that would tell us something about the reputation of webseries as a medium. To focus on, for instance, gender, I’d love to know for sure whether webseries tend to have higher female viewership than the average. And do those numbers change, based on the gender breakdown of the cast? In other words: Does the higher on-average representation of women in webseries make women more likely to watch a particular webseries than a comparable television show, even if they have similar ratios of women on-screen?

Now, I would love to do a full study on that, and maybe someday, I will. At present, though, I don’t have the time or the resources to take on that kind of research commitment. (As I mentioned above, it would require a lot of work even to design a study that avoided bias.) But while I wait for my schedule to free up, I have an interesting case study.

My friend Tom Pike is in the midst of launching a webseries, Personal Space, about a crew of astronauts whose therapy sessions are being broadcast, without their knowledge, as a reality show on Earth. (You can check out the Kickstarter here.) In addition to looking extremely cool as, like, a work of narrative, Personal Space has a feature that interests me from a statistical perspective: It maps extraordinarily well onto the traditional television series Battlestar Galactica. Personal Space and BSG share a similar genre (space-set sci-fi); have significant overlap in actors (Personal Space protagonist Nikki Clyne, second-billed Richard Hatch, and cast member Tahmoh Penikett are all BSG alumns); and, usefully for my purposes, have a similar gender breakdown among their main casts.

There are differences. Personal Space has yet to air, but its description sounds more surreal and less grim than Battlestar Galactica. And though Personal Space and BSG have similar gender ratios in their casts, Personal Space has a single, definitive female protagonist who is the face of the series (literally—her photo was the Facebook page photo until about two weeks ago). Battlestar Galactica, on the other hand, was a true ensemble, with Edward James Olmos billed first.

Nevertheless, it’s reasonable to think that, with the considerable on-paper similarities between the two series—and the fact that Personal Space is naturally marketed toward BSG fans, as BSG actors share posts and promotional materials—any significant difference in the gender ratio of the shows’ fanbases is likely to be partially due to the medium. (The fact that Personal Space has yet to air helps in this assumption; it reduces considerably the factors that could influence fan perception.)

The next step, naturally, is to study the gender breakdown of the shows’ fanbases. For a traditional television show like Battlestar Galactica, the most reliable way to do this would be to consult the Nielsen ratings. (I said most reliable, not actually reliable.) Unfortunately, detailed demographic information about Nielsen ratings isn’t publicly available. And of course, there would be no comparable numbers for Personal Space.

So instead, I looked at something a little more accessible: Facebook likes. Both Personal Space and Battlestar Galactica have Facebook pages, and Tom was kind enough to let me poke around in the Personal Space Facebook analytics, so I have reliable information about the demographics of Personal Space fans over time. The only problem is that a significant number of those likes will be from friends whom the cast and crew invited to like the page. (For instance, of the 841 current likes on Personal Space’s page, 245 happened on February 9, the day the page was launched—almost all of those are likely people who were invited.) Since friends of the cast and crew have incentive to like the Personal Space page regardless of their usual viewing habits, those likes probably aren’t informative. So I focused on people who liked the page between March 1 and April 15, when the page had been around for a while, and new likes were less likely to come from people who knew the cast and crew. That gave me a sample size of 128 fans, of whom 71 were women. So between March 1 and April 15, women accounted for approximately 55 percent of new fans of Personal Space.

(Interestingly, if you look at all 800 likes, from February 9 to April 15, 351 were from women—about 44 percent. That indicates that, as the Personal Space fanbase started to come more “from the wild,” so to speak, it became more female.)

I do not, alas, have the kind of relationship with Ronald D. Moore that would grant me access to the Battlestar Galactica Facebook analytics. So I went for the next-best option. Facebook allows you to search for users based on their likes. Unfortunately, the results aren’t random; the users you see are spit out by an algorithm that makes it more likely that you’ll see people you know, or who know lots of people you know. This biases the data in an unknown way—though, since I’m friends with a lot of female sci-fi fans, my guess is that it biases the data female. In order to minimize bias, I scrolled down the results until I found six users in a row with whom I shared at most one mutual friend. Then, looking only at users who had fewer than two mutual friends with me, and who had publicly shared their gender, I recorded the gender of 128 Battlestar Galactica fans. Of those 128 fans, 32 were women—25 percent, exactly. To verify that that number wasn’t wildly off, I compared it to the demographics of IMDb users who have rated Battlestar Galactica, demographics which, to my surprise and delight, IMDb makes public. Slightly over 13 percent of the 117,734 people who have rated Battlestar Galactica on IMDb are female.

So even if my Battlestar Galactica numbers are biased, they’re probably biased female—and if my Personal Space numbers are biased, it seems reasonably likely that they’re biased male. Which means that there is a significant difference in the gender breakdown of the shows’ fanbases. Personal Space has many more female fans than Battlestar Galactica.

Obviously, this is just a case study, not a full statistical study. I can’t even construct a distribution, let alone prove causation. And of course, to really talk about diversity in audiences, you’d have to look at factors other than gender—race, sexuality, age, etc.—that are much harder to get reliable public data for. But it’s a starting point, and something to think about. Hopefully, as the number of webseries—and the number of people watching them—continues to grow, more people will start to study this medium. I, for one, can’t wait for more data.

No comments:

Post a Comment