3 ms·
Well-designed survey results are objective and reproducible. There are margins of error as with all measurements, and maybe these margins are wider than in some
by ridaj 5y ago
Well-designed survey results are objective and reproducible. There are margins of error as with all measurements, and maybe these margins are wider than in some fields and narrower than in others, but asking people for their opinions isn't necessarily bad science just because it happens to involve opinions.
- simonh 5y agoRight which is where attempts to confirm the thesis come in, and as the article shows it looks like Gerber's conclusions were based on a heavily loaded question that severely biased the answers.