home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!IRLEARN.UCD.IE!RCONROY
- Message-ID: <EDSTAT-L%92111309081806@NCSUVM.CC.NCSU.EDU>
- Newsgroups: bit.listserv.edstat-l
- Date: Fri, 13 Nov 1992 14:03:13 GMT
- Reply-To: Ronan M Conroy <RCONROY@IRLEARN.UCD.IE>
- Sender: "Statistics Education Discussion" <EDSTAT-L@NCSUVM.BITNET>
- From: Ronan M Conroy <RCONROY@IRLEARN.UCD.IE>
- Subject: Teaching interpretation of correlations
- Lines: 19
-
- The discussion about what correlations measure and how they should
- be interpreted beings me to an amusement that I've started
- perpetraing in stats sessions. Every time students get a
- scatterplot, I ask them to give it marks out of 1 - 1 for
- a perfect line, 0 for a complete feck. It is only by doing
- this repeatedly that they get any idea of what a correlation
- actually looks like, and can begin to picture a correlation?1;2c
- and get some idea what the number means. I believe that
- most people, on seeing a correlation coefficient, get no mental
- picture at all. In fact, I believe them to be almost valueless
- in their ability to communicate with the average reader of a
- scientific paper. Anyone else agree?
-
- Ronan Conroy EMAIL RCONROY@IRLEARN.UCD.IE
- Lecturer in biostatistics Royal College of Surgeons in Ireland
- and computing and stuff Mercer Building, Dublin 2.
- (353)-(1)-780200 [fax:780934]
- ---------------------------------
- "Sutor ne ultra crepidam judicaret"
-