By far the most popular post on my blog (despite its length) was my post late last year about social network demographics (see “Age *Does* Matter: On the Demographics of Social Networks”). Among the phenomena it discussed was the clearly exponential increase in Facebook usership across age groups (younger = faster). Here’s the chart:
In his posting A Profile of Online Profiles today to the excellent bi-weekly By the Numbers blog, Charles M. Blow, the New York Times’s Visual Op-Ed columnist, posted a chart that looks startlingly similar:
How similar are these curves? Pretty damn similar! That’s remarkable, given that I was originally building my data from students who graduated college (or at least voluntarily associated themselves with a college class), and the Rapleaf data on which Blow’s chart is drawn comes from the broader population. [Thanks to my friend Michel Floyd for pointing out that I had previously glossed over this signficant methodological inconsistency.] Nevertheless, the similarity is striking—and thanks to a little quick-and-dirty overlayment:
They’re both clearly exponential. But there is a slight difference in the “middle” of the curves. It’s possible that the “real” data is a bit steeper on the tails (leptokurtic), that the demographics have shifted in the past few months, but in all likelihood the reason for the slight discrepancy in the middle is that my data were from a different sample (college graduates versus general population) and prone to certain systematic errors that RapLeaf’s aren’t. (For details, see the methodological notes to my prior post.)
Blow’s data, by the way, are based on data published this summer by my friend Auren Hoffman‘s company, RapLeaf, by the way. RapLeaf has a spectacularly rich database on social network profiles and has the unique ability to track users from site to site.