A misunderstanding of social media is driving media elites to keep pushing an easily disprovable stereotype
by Keith A. Spencer
Part 3 - Angry people and angry brands
But the demographics of Twitter's user base only say so much about the site's distorted commentariat. There's also the question of how people behave online, and why they behave so differently than they do in real life. There is a psychological reason why even very nice people are more likely to behave like assholes online. It is called the online disinhibition effect, and it is a big source of misery from pundits who do not understand it.
The combination of three factors — the anonymity and pseudonymity of being online, the lack of accountability, and the indirect nature of online communications — make it so that online communication is dehumanizing, and often cruel.
Demographics and "real" users aside, Twitter — like most social media sites — has a huge number of accounts that aren't even individuals. A great deal of Twitter users are instead are brands, spam accounts or bots who behave like actual people.
Because of this, getting in arguments with "people" on Twitter — or even just seeing Twitter as the so-called public sphere — is akin to arguing politics with a clown in a funhouse mirror. It is so heavily distorted — by corporate PR and marketing, by the way that people behave differently online, and even by powerful bad actors (whether state or individual) who can wield Twitter armies quickly and easily — as to be effectively useless as any sort of gauge of public opinion.
It is a terrible place to gauge human behavior, or make broad pronouncements of what humans are like. And it's an even worse place to get a sense of a politician's support base.
I have a modest proposal for my peers in the journalism world: I would like to propose that anyone writing about a Twitter "mob" of any political ilk be required to include the previous paragraph in an asterisk at the bottom of their story. We should all be forced to include a disclaimer to clarify that it is impossible to make any kind of quantitative assessment of human behavior on Twitter because of how deeply skewed it all is — by hackers, PR professionals, paid influencers, intentional government or corporate misinformation campaigns, and the way the online disinhibition effect makes people act.