Twitter reactions not representative of public opinion

The same way you can prove Pew (or other polling services) are legitimate.

Who are they calling? Are they spanning the entire spectrum of Americans for any particular issue they want opinions on?

Do they divulge a list for the people they contact for a poll? Do they explain the people on that poll? How do they know the people in that poll are giving honest, truthful answers/opinions? And so on and so on.

I would trust the underlying meaning of a tweet because after a while you start to understand the person behind their tweets. At least it's out there in the open for the world to see.

 

Yes, I trust the underlying meaning of a tweet.

But a tweet is quite different than a collection of tweets purporting, at least impliedly, to represent public opinion.

 

Agrees.

 

...where credit is due. Pretty solid hammer

 

Well, you'd have to explain that one being that I missed it. Did he pronounce some victory because of Twitter?

 

Leitch wrote about this during the Olympics. 100 percent -- or so -- of people on Twitter hated the tape-delay. But the majority of the world was probably fine with it. Someone working at a job where they're not in front of a computer or a TV during the day is happy to come home at night and watch Michael Phelps win gold even if it happened six hours earlier.

 

Sure, we read results on Twitter (and some of that no doubt spilled over to Facebook) but I still enjoyed watching the events later that night. It's one thing to read a result in real-time, another to be able to watch it too.

 

Exhibit No. 3,977,210 in "Journalists don't understand math."

 

I don't understand. Do you think lantaur is being sarcastic and you are chiding him because polls really are accurate?

Is he wrong on the actual number of Americans?

Help me out here.

 

Truly random, relatively small samples are highly accurate in representing much larger populations. It's counterintuitive to people who don't understand statistics, but it's true.