Conversation

Aella
@Aella_Girl
critiques i get from ppl who don't know stats: "Selection bias! Internet surveys aren't 'studies'! Just Aella's Audience!" from ppl who do know stats: "You should try [different new statistical method] instead, would account for [obscure math thing I need to go learn now]"
35
458
Low T-Pain
@bifidis
Replying to
@Aella_Girl
I’m sorry but you need to touch grass if you think you can draw population-level conclusions here. There’s just absolutely no way that these surveys are being completed by even 40th percentile cultural-norm-following Americans. Begging u to show evidence of the statistical power
1
5
Low T-Pain
@bifidis
Replying to
@bifidis
and
@Aella_Girl
I accept any kind of analysis that you’d present as “of the 500k people who completed my survey, there was a statistically significant relationship between X and Y” but not any broader claim about any other population. That can be statistically valid
1
2
Low T-Pain
@bifidis
Replying to
@Aella_Girl
Literally the tweet that I replied to. You’re dismissing selection bias, which is just another way of saying your surveys aren’t representative
1