Conversation

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]"

458

James, expert on every trending topic

@ksqueed

Jan 31

Replying to

@Aella_Girl

The EXCEPTIONALLY high IQ’s associated with your recent survey results demonstrate that a great deal of selection bias (or self-delusions among respondents) exists in your data.

Aella

@Aella_Girl

Replying to

@ksqueed

among self reported iq yes! my guess is the effect will hold regardless of the population you run thar question set in

2:20 AM · Feb 1, 2023

305

Views

James, expert on every trending topic

@ksqueed

Feb 1

Replying to

@Aella_Girl

Respondents will self-select to some degree as typically a low percentage of the sample responds to the survey. But if you sample from a narrow category, your results are not considered representative beyond that category (which is really a study design issue vs. a stats issue).