[R] Difference in ANOVA results - R vs. JMP/Minitab

John Fox jfox at mcmaster.ca
Wed Nov 19 15:57:40 CET 2003

Dear Nirmal,

At 09:12 AM 11/19/2003 -0500, Nirmal Govind wrote:
>Thank you very much John and Sundar for your quick responses.
>>  From your description, it's not possible to tell what you did or 
>> exactly what happened -- either in R or in Minitab (and JMP).
>John: in R, I fit the model using lm exactly as you did.. however, for the 
>ANOVA table, I used summary(lm) and also tried aov(lm) but not anova(lm) 
>... and it looks like therein lies the problem.. the output of these differ.
>What's the difference between aov, summary and anova? And when should one 
>use one as opposed to the other? It's hard to tell from the help for these 

Briefly, as ?aov mentions, aov() calls lm() -- the difference is in the 
results produced by generic functions like summary(), print(), etc. For 
example, summary.lm() produces a table of coefficients, std. errors, etc., 
while summary.aov() produces an ANOVA table. One would not normally call 
aov() with a linear-model object as an argument (though this works). 
Applied to a linear-model object, summary() produces coefficients, etc. (as 
mentioned), while anova() produces a (sequential) ANOVA table. This seems 
apparent to me from the output.

More generally, it probably makes sense to read introductory material about 
R -- such as the introductory manual that comes with the software or one of 
several books (some freely available on the Internet) that describe its use 
(see under documentation at <http://www.r-project.org/>).


John Fox
Department of Sociology
McMaster University
Hamilton, Ontario, Canada L8S 4M4
email: jfox at mcmaster.ca
phone: 905-525-9140x23604
web: www.socsci.mcmaster.ca/jfox

More information about the R-help mailing list