To repeat my previous comments, but to clear up a few confusions, the most recent survey on expert’s opinions on intelligence is based on 228 participants. The percentage participation rate is lower than in 1984. The academic climate may be more hostile, the questionnaire too long, the nature of the questions not appealing, or other reasons which might be described in the eventual paper (which should also give the precise number of replies for each question).
However, this is the best estimate we have at the moment, short of government surveillance of emails and private conversations…….. Now, there’s a research methodology which should get us a good sample size!
The lower the response rate, the higher the influence of self-selection bias.
ReplyDeleteSure, but which way?
DeleteOh, I would have thought the safe, politically correct, "nice" way, of course!
ReplyDelete