home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1992 #23 / NN_1992_23.iso / spool / bit / listserv / statl / 1729 < prev    next >
Encoding:
Text File  |  1992-10-13  |  1.1 KB  |  25 lines
  1. Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
  2. Path: sparky!uunet!zaphod.mps.ohio-state.edu!sdd.hp.com!wupost!darwin.sura.net!paladin.american.edu!auvm!SEQ.HULL.AC.UK!R.A.REESE
  3. Via:         UK.AC.HULL.SEQ; 13 OCT 92 10:55:11 BST
  4. X-Mailer:    ELM [version 2.3 PL11]
  5. Original-Sender: R.A.Reese@uk.ac.hull.seq
  6. Message-ID: <STAT-L%92101306034743@VM1.MCGILL.CA>
  7. Newsgroups: bit.listserv.stat-l
  8. Date:         Tue, 13 Oct 1992 06:03:47 -0400
  9. Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
  10. From:         Allan Reese <R.A.Reese@SEQ.HULL.AC.UK>
  11. Subject:      Re: Need help with PCA
  12. In-Reply-To:  <received 13 Oct 92 03:33:34 via COM>; from
  13.               "Meredith_Warshaw@com.FRANKSTON" at Oct 12, 92 10:25 pm
  14. Lines: 9
  15.  
  16. > concerns about "data snooping".  On the other hand, with only 15 animals,
  17. > some of which have
  18. > missing data,
  19.  
  20. W.E.Deming makes a good point in "Sampling for Business Studies" (1960)
  21. [query the title but it's something like that] that very small random
  22. samples are inherently unreliable, so that for small studies it will
  23. often be preferable to choose a "judgemental sample" in which case
  24. probabilities are not applicable for inference.
  25.