Article View: sci.math.stat
Article #7ANOVA interaction terms?
From: Darrell Roy Hougen
Date: Tue, 27 Oct 1992 17:07
Date: Tue, 27 Oct 1992 17:07
16 lines
1206 bytes
1206 bytes
My wife is trying to do some ANOVA on her mouse tumor data using
SYSTAT. The program automatically generates a model of the form
F = Const. + A + B + A*B
However, the model can be changed by the user. What I am wondering,
not knowing much about ANOVA, is what purpose does the A*B term serve?
My undergrad Stat. book doesn't say anything about such terms. Could
anyone recommend one or more (so I have a chance of finding one) texts
on the subject? Right now, the extra term looks pretty ad hoc. Is
there a sound reason for using a term of that form? Why not use A^2
and B^2 too? That would be a complete second order model.
Thanks in advance.
Darrell
Technical Details
Message-ID:
<1cjsvkINN5c0@roundup.crhc.uiuc.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!ogicse!uwm.edu!ux1.cso.uiuc.edu!roundup.crhc.uiuc.edu!focus!hougenShow Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!ogicse!uwm.edu!ux1.cso.uiuc.edu!roundup.crhc.uiuc.edu!focus!hougen From: hougen@focus.csl.uiuc.edu (Darrell Roy Hougen) Newsgroups: sci.math.stat Subject: ANOVA interaction terms? Summary: Why use them? Keywords: ANOVA Message-ID: <1cjsvkINN5c0@roundup.crhc.uiuc.edu> Date: 27 Oct 92 17:07:00 GMT Article-I.D.: roundup.1cjsvkINN5c0 Organization: Center for Reliable and High-Performance Computing, University of Illinois at Urbana-Champaign Lines: 16 NNTP-Posting-Host: focus.csl.uiuc.edu