this post was submitted on 21 Jun 2024
381 points (96.6% liked)

Data is Beautiful

1138 readers
8 users here now

Be respectful

founded 4 months ago
MODERATORS
 
you are viewing a single comment's thread
view the rest of the comments
[–] benjhm@sopuli.xyz 22 points 4 months ago (6 children)

That's interesting. I wonder whether those 6519 surveyed are representative of whole population, or of people who anyway online a lot. It’s seems there was an inflection around 2012 - what happened then ? The curve ends during covid lockdowns, wonder whether deflected since ?

[–] alicemichelle@lemmy.today 6 points 4 months ago (2 children)

Based on the one class I took in college about surveys and mass comm I’d say that’s a good sample size (assuming they were chosen at random). Most political polls survey about 1500 people with 90%+ accuracy

[–] benjhm@sopuli.xyz 5 points 4 months ago (1 children)

But how, practically, do you choose any sample "at random" nowadays ?
Especially if trying to avoid a bias towards (or away from) online people ?

[–] alicemichelle@lemmy.today 4 points 4 months ago

Great question! Back in the day we would pick names at random in the phone book so my info is pretty outdated

load more comments (3 replies)