> > You are bit by bit dragging me out on thin ice here (statistics and > probability) which is fine, so long as you are prepared to rescue me. > > I think, as a matter of practice, that the strength of an inference is > determined *a priori* when you define your population and select your > sample size. > > Does that sound right? > The ice is as thin for me as for you but I would think that the probable maximum strength of an inference is determined by the nature of the sample (that can be measured within just the sample). So we can only make very weak inferences concerning life on other planets, because we have a sample size of one. But if the first exoplanet we find with life on it has only hominids, then an inference that 'dominant' lifeforms can only be hominids would appear to double in strength but might not actually be stronger than before at all if it turns out just to be luck. I may revise this opinion upon further rumination, though, as I feel like my analytical skills are not at their strongest currently. -Arlo
============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
