lingo.lol is one of the many independent Mastodon servers you can use to participate in the fediverse.
A place for linguists, philologists, and other lovers of languages.

Server stats:

69
active users

Dr. Angus Andrea Grieve-Smith

Welcome to everyone who's just joining or rejoining Mastodon!

Here's your periodic notice that it is statistically impossible to get a representative sample with a Mastodon poll. The most reliable thing you can find out are absolute numbers ("34 people think Aaron Neville is underrated!").

Please feel free to create, vote in and boost polls for fun. But please don't mislead anyone into thinking that boosting will achieve a "better sample size." It will not!

@grvsmth A representative sample? In this economy?

@grvsmth (I know this is an old post but) why not? Boosting spreads the poll to new people and increases the sample size, larger sample size is better, no?

@elduvelle No, because it's not actually a sample. I'll post a link soon.

@grvsmth I took some stats in Uni. Random samples it is not.

@grvsmth Just to be clear, everyone knows that Aaron Neville is underrated.

@grvsmth I would, but as you have pointed out, it is impossible on Mastodon.

@glennsills No, you'd have to poll a representative sample of the relevant population!

@grvsmth To be fair, you could only determine if everyone thinks Aaron Neville is underrated if you polled everyone, even though, we all know it is true.

@glennsills Yes, that is the only way to be sure -although there is uncertainty even with that. But we can use sampling to verify unanimity to within an acceptable margin of error!

@grvsmth But never on Mastodon, because of the way it is organized. Which gets us all the way back round to your original point.😀

@glennsills I suppose it's possible to have 100% of the population subscribed to a single Mastodon account, but I don't see it happening any time soon...

@grvsmth Even then the population would be skewed to the people with Mastodon accounts, which is far from universal.

@glennsills That's what I mean, if everyone in the world joined Mastodon and subscribed to a single account, we could have that account invite them to take a survey.

But if 100% of the population thinks Aaron Neville is underrated, then who do they think underrates him?

That's a good point Dr. Angus which just goes to show you how incredibly outrageous your original "for instance" was... 😃