Correlation with categorical variables

Article View: bit.listserv.stat-l

Article #3558

Back to bit.listserv.stat-l Previous Next

#3558

From: R. I. Beekun
Date: Sun, 26 Jan 1992 21:39

12 lines
1195 bytes

I have a categorical variable that I wish to correlate with a number of
continuous variables.  The categorical variable has four categories (i.e. a
firm can be either in category A or B or C or D).  I have been looking
up my statistics reference books, but cannot get a clear answer.

Which type of correlation is appropriate for this situation? Would
Kendall's tau or Spearman's coefficient be more appropriate than
Pearson's coefficient?

Any advice would be much appreciated.

beekun@unssun.nevada.edu

Technical Details

Message-ID: <9201262139.AA11010@unssun>

Path: arc1-ca.pugleaf.net!sparky!uunet!paladin.american.edu!auvm!UNSSUN.NEVADA.EDU!BEEKUN

Show Full Headers

Complete Article Headers

X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: f5efb,4154c82f052baad2,start
X-Google-Attributes: gidf5efb,public
X-Google-ArrivalTime: 1992-01-26 14:33:13 PST
Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
Path: arc1-ca.pugleaf.net!sparky!uunet!paladin.american.edu!auvm!UNSSUN.NEVADA.EDU!BEEKUN
Message-ID: <9201262139.AA11010@unssun>
Newsgroups: bit.listserv.stat-l
Date:         Sun, 26 Jan 1992 13:39:43 PST
Sender:       "STATISTICAL CONSULTING" <STAT-L@MCGILL1.BITNET>
From:         "R. I. Beekun" <beekun@UNSSUN.NEVADA.EDU>
Subject:      Correlation with categorical variables
X-To:         stat-l@vm1.mcgill.ca
X-cc:         methods%rpiecs.bitnet@ubvm.cc.buffalo.edu