home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1992 #26 / NN_1992_26.iso / spool / bit / listserv / statl / 1961 < prev    next >
Encoding:
Text File  |  1992-11-13  |  1018 b   |  22 lines
  1. Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
  2. Path: sparky!uunet!paladin.american.edu!auvm!UA1VM.BITNET!EJOHNSO3
  3. Message-ID: <STAT-L%92111308544382@VM1.MCGILL.CA>
  4. Newsgroups: bit.listserv.stat-l
  5. Date:         Fri, 13 Nov 1992 07:42:44 CST
  6. Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
  7. From:         EJOHNSO3@UA1VM.BITNET
  8. Subject:      Test for assumptions
  9. Lines: 11
  10.  
  11. I would like to know how regularly people on this list test for assumptions.
  12. For example, when a t test assumes a normal distribution and equal variances,
  13. how often do you run a Lilliefor test for normalcy and an F test for the dif-
  14. ference of means? I'm told that it should aways be done, but these tests don't
  15. seem to be a standard part of statistical packages such as SAS, nor do I recall
  16. having seen these tests being mentioned in the methodology of experimental re-
  17. search. Comments, please.
  18.  
  19.                           --Ed Johnson
  20.                             Dept. of Mass Comm.
  21.                             University of Alabama
  22.