Correlation does not prove causation, as any anonymous Internet commenter will tell you.* That said, here's some very interesting data on Twitter usage reported on by Business Insider, which shows a sharp dropoff starting in August 2014. (See chart above.)
What happened then? Well, for one thing: "Gamergate" first became a pseudo-movement on Twitter in August 2014, with thousands of anonymous anti-feminist reactionaries trolling and abusing anyone they blamed for introducing feminist/social justice issues into games on Twitter. (Especially women.) Business Insider attributes the spike and subsequent drop to the end of the World Cup, but here's the thing: Media coverage of Gamergate and the abusive behavior Twitter also starting going up in August 2014, which is a great way to discourage most people from using the social network at all (especially women).
This correlation was first pointed out by Redditor "/lastres0rt", who puts it this way:
Ordinarily I would think that because of the World Cup mention (could just be a coincidence), but note the dramatic decline in Twitter activity after that peak.
It didn't merely return to normal, it plummeted to HALF its original, stable baseline before July / August 2014, and continues to be declining. That implies something a bit more serious is not only going on, but continuing to occur.
(Also, remember that GG drove off several high-profile celebrities and other "high value" users -- users which were probably not only being active on Twitter, but also attracting new blood and encouraging discussion on those networks. Their absence is likely being measured as well.)
To take one example, actress Anna Kendrick, who doesn't even work in game development, was swarmed by GamerGaters. At minimum, we can definitely say Gamergate didn't help growth in Twitter usage -- and certainly not from Gamergaters themselves. As I first pointed out in September 2014, actual people associated with the term seemed well under 10,000 furiously furiously Tweeting. And it highlights former Twitter CEO Dick Costolo's point that the company sucks at dealing with abuse. Sucks so hard, it seems, that it's been hemorrhaging users ever since -- to the point where the whole company is now in jeopardy.
*Correlation does not prove causation, but referring to correlation on the Internet does in fact cause people to post "Correlation does not prove causation!" on the Internet.
Comments