“There’s no hard-and-fast data that quantifies… but we know”

This pernicious phrase reared its ugly head in a CNET piece about Facebook and teens:

There’s no hard-and-fast data that quantifies Facebook’s teen problem. But we know — from observing teens, talking to parents and analysts, and from a few company statements — that age doesn’t become Facebook with this group.

There is a reason we do research, surveys and studies. Because we don’t know and cannot make definitive statements without hard data and evidence. Saying that there isn’t hard data but we know anyway is a trope used by journalists trying to justify writing about something that they feel is true, even though they have no data to support it. Or as a way to justify writing some provocative that will get attention, even though the evidence is not there.

The core issue is the “but we know.” There is nothing in this CNET piece that suggests they know anything. There are beliefs supported by circumstantial evidence, some of it mildly compelling, some of it not.

I have no issue with writing an opinion piece that is based on anecdotes and circumstantial evidence to ask a question, to begin to posit that studies and date are needed. But this piece doesn’t ask a question. It states opinion as fact, as dangerous of a journalistic sin one can commit.

Talking to some teens or some of your friends’ children doesn’t count as research. You don’t know if they are representative of the larger population, thus you can’t generalize this information to the population at large. Maybe it is true that in San Diego, where the author Jennifer Van Grove is based, teens are leaving Facebook for other social networks, but what does that tell us about rural Kansas or Cleveland or Washington, DC? Or even neighboring school districts from the one she lives in?

We are also given no indication of how many teens she observed, how she observed them, who she talked to, how many people she talked to, where those people live, how she got ahold of these people to talk, etc. In short, she provides no evidence that she collected her data in any sort of scientifically relevant manner

But it gets worse.

The core thesis to this poorly researched piece is that there isn’t hard data, but that through anecdotes and conversations with unrepresentative samples, we know the truth. But there is some data to compare this thesis to.

Garry Tan of Y Combinator surveyed 1,038 people about their social networking usage, broken down between 538 responses from users aged 13-18 and 492 responses from users 19-25. I haven’t seen the methodology of this survey, so I can’t speak to its scientific veracity and its almost certainly not of the caliber that I could use in my research, but it is still better than a reporter — who is not a trained ethnographer — observing some teens and talking with some parents and reading tea leaves.

Tan’s data shows that 55 percent of 13-18 year-olds surveyed use Facebook. That is surpassed by Tumblr, which has 61 percent penetration. Instagram, the social network that Grove said was stealing Facebook’s thunder, is used by 21 percent of users surveyed (and it just passed 100 million users, less than a tenth of Facebook’s, meaning that it would need to be wildly overrepresented amongst teens to be more popular with teens than Facebook).

Snapchat, another social network that as cited by Grove as being an issue for Facebook was even less used at 13 percent (and may have under 5 million monthly users). When I first read the title of Grove’s piece, I assumed she was talking about Tumblr because it is one of the 10 most popular sites in the U.S., and half of its user base under 25. But no, Tumblr a site that by many measures is doing quite well with younger audiences, was mentioned once in Grove’s piece.

This is why anecdata, which Grove and many journalists fall prey to, is useless. I have no doubt that within certain pockets of users, Instagram and Snapchat are huge and that Facebook is passé. But when we talk about trends, we need good data that is generalizable across a population.

If you were to go to a high school, you would find different things popular within certain cliques of students. What makes Groves’s analysis so specious, is that if she observed a certain clique of students, it is possible that Instagram was all the rage among their friends, but not even generalizable across that one school.

I don’t doubt that Facebook is no longer the cool social network. It is used by more than one billion people every month, has amazing market penetration and is popular with all age groups. It’s not hot, new or cutting edge.

But Facebook is the standard bearer. It has proven itself over nine years of use, a long time in the land of social networking. It may never again be the hot, young social network. Just as The Rolling Stone or U2 may never again be the hot, young rock band. But Facebook may prove to be something much bigger — enduring.

Beyond this, none of this data or anecdata shows that teens won’t become Facebook users as they age. Facebook was always a social network for an older audiences than middle and high school students (it was originally college only). The real question will be whether or not people don’t age into Facebook, and we are a few years away from knowing that.

I don’t disagree that Tumblr, Instagram, Snapchat and other newer networks are better at anonymity than Facebook, which is something that appeals to teens. Grove argues that lack of anonymity and the presence of parents and grandparents causes Facebook to not be cool with teens. But anonymity is also something that causes weak ties.

What makes Facebook enduring, and what makes it something that has appealed to older users, is that Facebook users tend to have their strong ties on the network — family and close friends. The presence of strong ties is the impetus for many people to join Facebook in the first place.

I’ll delve deeper into the difference between strong and weak ties in the future, but it is an important concept to understand when it comes to analyzing social networks. The average person has at most 10 strong connections in their life, and these are connections that often last a lifetime.

13-18 year-olds are just beginning to forge strong ties and are often antagonistic to future strong ties (family). So, I would be careful when comparing social networks that appeal to weak ties to those that appeal to strong ties.