But you could say that about anything including the opposite.
I see a bunch of people who say that "CEO-killing is wrong. Therefore they must be in an internet bubble." I think we should both admit that the consistency of a message isn't actually a good signal for "bubbleness" and that something like randomized polling on personal beliefs and perceptions or a similar study actually would be.
Bubbleness isn't about the viewpoint, it is about the difference between bubble perception vs global perception.
If you read reddit/twitter, a common statement was something like "The police will never be able to find him - no one will cooperate". "Must be hard to find someone when there are 150million suspects". Basically, treating him like a modern day robin hood.
When, back in real life, the news of the day seems to be that he was caught at a McDonalds after two random employees noticed him and called the police on him.
If you want to make a statement about bubbles, then you have to ground it in global perception which is operationalizable and empirically verifiable - speculation isn't epistemologically responsible.
I'm not particularly worried about verification or epistemologically responsibility when something is manifestly obvious. The same reason I'm not jumping out of an airplane without a parachute even if I haven't read a study on the relative effectiveness of parachutes vs no parachutes when jumping out of an airplane.
Sorry, I hate to burst your bubble, but social facts ESPECIALLY need to be tested before being assumed because humans are particularly susceptible to typification, legitimization, and reification.
Sure, it's possible to acquire my magnificent skills, by just going out and interacting with a wide swath of people in the real world.
Even, instead of thinking that you're getting any kind of signal about society when you read the same meme comment on Reddit/Twitter or whatever, just imagine you actually have no idea, even if you've read the 5000th tweet expressing the same idea