In Which A Study Is Actually a Joke
Ed. Note: I started writing this in August, and as such, it is dated. However, I feel that this topic is still important, so I finished it and decided to publish it
Well, I’d never thought I’d see the day. The news media has taken a study and completely botched reporting it. Via PCMag.com, apparently there’s some sort of studying making rounds on the internet and on the cable news channels (which I don’t watch, thank Zoroaster).
Look, I think that the reporting on this study is so ridiculously craven and catered to the pearl-clutching mother demographic that the only real word I can think of to describe it accurately is “GGAAAAAAAHHHRRRRRFFFFFFFAAAAHHH!” That’s right.
There’s this narrative going on that “teh Facebook” causes people to do stupid things. The study, as reported by the media, tries to paint a causal relationship between Facebook and teenage drug use (among other things). This is why we can’t have nice things: the study actually only looks at correlation and because of the nature of the study, it is simply not valid to say that Facebook causes drug use. It just so happens that teenagers who use the social networking website Facebook also have a social life. Shocking, I know.
In essence, the study examined two variables:
- Social network use
- Drug, tobacco, and alcohol use
While study found high correlations between social network use, such as Facebook and MySpace, and access to drugs and drug abuse, the study made no claims that these networking sites caused drug use. This is an important distinction to make, especially in light of the way that the media has been reporting it. It is, simply, incorrect to report that the study claimed a causal connection. There are many variables that exist outside of the purview of the study which most likely contribute to the problem more than Facebook or MySpace. Teens who do use alcohol and drugs most likely have live social circles that exists alongside their technological counterparts. It seems unlikely, then, that these networking sites are a large factor, because these social circles are, I believe, the primary enablers and motivators of this behavior.
Let me explain in greater detail why the correlations found in the study shouldn’t be used to establish a causal connection. The first set of data reports that teens who use social networking sites are more likely to have used tobacco, alcohol, and marijuana (p. 13). The study also reports that teenagers who use social networking websites experience higher levels of exposure to pictures of tobacco, alcohol, and marijuana being used by their peers (p. 14). While this may be true, it doesn’t prove that social network use causes teenagers who view those picture online are being influenced by those pictures, and not other factors. For instance, if their main group of offline peers are the people who are displayed in those pictures, the influence may come from old-fashioned peer pressure.
This is why it’s so important to have a skill set to interpret the results of these studies. I really wish I didn’t have to say this, but the way that the news media reports studies like this is just not usually very accurate. I can’t say if this is because news outlets lack the ability to determine the very real differences between correlational research and experimental designs (which often make claims about causation), of if it’s a strategy to hook readers and get more eyes reading the article.
Either way, it’s highly misleading.