home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!wupost!darwin.sura.net!paladin.american.edu!auvm!SWIRL.MONSANTO.COM!GIBES
- Message-ID: <9207271741.AA23616@tin.monsanto.com>
- Newsgroups: bit.listserv.stat-l
- Date: Mon, 27 Jul 1992 12:41:32 -0500
- Sender: "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
- From: Kernon Gibes <gibes@SWIRL.MONSANTO.COM>
- Subject: RE: Is it good agreement ?
- X-To: stat-l@vm1.mcgill.ca@tin.monsanto.com
- X-cc: GIBES@tin.monsanto.com
- Lines: 17
-
- Regarding P. S. Gill's "good agreement" question, I have some
- non-statistical questions:
-
- As stated, the point of submitting the questionnaire to 5
- experts was to "test" the questionnaire. But, I'm wondering,
- *IF* the objective of the questionnaire is attitude assessment,
- what could be inferred from agreement OR disagreement among N=5?
- These 5 people may very well tend to have the same assessments
- as to appropriateness of the actions presented insofar as they
- are, at best, a (non-random?) sample of the population of
- experts. Alternatively, if they disagree, does this imply that
- the questionnaire somehow fails to pass this "test" or merely
- that the questionnaire is very good at presenting situations for
- which no wide consensus exists?
-
- Kernon Gibes
- Internet: gibes@swirl.monsanto.com
-